Exposure to particulate matter (PM) has been associated with increased mortality and hospital admissions due to cardiovascular and respiratory disease , including influenza-related mortality . We and others have previously shown that exposure to DE increases the susceptibility for viral infections [15, 16], such as influenza [17, 18]. NK cells play important roles during antiviral host defense response by killing virus infected host cells and by initiating an immune response via the release of cytokines [3, 4]. Despite the important roles NK cells play in respiratory immunology and host defense, the potential impact of air pollutants, such as DE, on NK cells function has not been thoroughly investigated. Our results indicate that in the context of stimulation with the viral mimetic pI:C, DEP decreases markers of cytotoxic function, such as CD16 expression as well as the expression of granzyme B and perforin, and suppresses the ability of NK cells to kill target cells. Thus, impaired antiviral host defense responses seen after exposure to DEP may be related to the direct effects these pollutants have on NK cell function.
In addition to producing cytokines , direct killing of (virus-) infected cells and transformed tumor cells are functions of NK cells . Reducing the cytotoxic potential of NK cells by exposure to environmental stressor, such as particulate matter, could affect the ability to fight viral infections. We have recently demonstrated that in nasal passages of human volunteers, NK cells comprise a significant portion of the mucosal lymphocytes .
These studies also demonstrated that in the context of viral infection, markers of NK cell cytotoxicity were reduced in smokers as compared to non-smokers , which was associated with increased viral replication in a similar cohort . The data presented here demonstrate that DEP suppresses the cytotoxic potential of NK cells. Our previous study conducted in human volunteers exposed to DE prior to inoculation with live attenuated influenza virus (LAIV) indicated that DE exposure increases quantity of virus in nasal secretions . DE-induced suppression of NK cell cytotoxicity, as shown here, could lead to reduced killing of virus-infected host cells and facilitate enhanced virus replication in the nasal mucosa.
NK cells use three different mechanisms to carry out their cytotoxic effector function: 1) Release of cytotoxic mediators (granzyme B and perforin), 2) Fas-Fas ligand mediated pathway, and 3) cytokine dependent pathway (cross-linking of TNF and TNF receptor) . Our data show decreased gene expression as well as protein levels of granzyme B and perforin due to DEP exposure (Figure 1), which correlated with a reduced ability of NK cells to kill target cells after DEP exposure (Figure 4). In addition, our data presented in Table 1 indicate no significant effects of DEP on pI:C-induced TNF-α release by NK cells, suggesting that the TNF/TNFR (mechanisms 3) does not play an important role in mediating the DEP-induced suppression of the NK cell cytotoxicity. We cannot completely exclude Fas-Fas ligand mediated pathway playing a role in the DEP-induced suppression of NK cell cytotoxicity. Even though 7-AAD used in the cytotoxicity assay here stains target cells after entering dead or damaged cells by entering through pores in the cell membrane, thus favoring necrotic cells, we cannot rule out apoptotic cell death of target cells in the cell-mediated cytotoxicity assay used here.
Exposure to DEP could affect the granzyme B/perforin mediated cytotoxicity at two levels. First, the expression of granzyme B and perforin can be regulated transcriptionally. Various signaling pathways and transcription factors can be involved, such as Ikaros, core-binding factor (CBF), activator protein (AP) -1, nuclear factor kappa B (NF-κB), janus kinase (JAK) 1, signal transducer and activator of transcription (STAT) 3 and 5 [25–27]. Studies conducted in bronchial epithelial cells have demonstrated that exposure to DEP activates AP-1, NF-kB, and STAT3 [28, 29]. Second, the lysis of granzyme B and perforin depends on the intracellular movement of granzyme B and perforin containing vesicles to the target cell. Phosphatidylinositide 3-kinases (PI3K) plays a pivotal role in regulating this intracellular movement . PI3K and NFκB are known to be affected by DEP exposure  and may be involved in the DEP-induced suppression of NK cell cytotoxicity. Thus, DEP may impair NK cell cytotoxicity, and more specifically granzyme B and perforin formation and release both transcriptionally and post-translationally.
In addition to cell-mediated cytotoxicity, NK cells play important roles during innate immune responses and provide a bridge to adaptive immunity by being a significant source of cytokines . In particular, NK cells have been shown to be important sources for IFN-γ, TNF-α, IL-1, IL-2, IL-4, IL-5, IL-10, IL-13, IL-8, and IL-12 [32–35]. We tested the production of these cytokines by NK cells and demonstrated that most were enhanced by stimulation with pI:C alone or pI:C+DEP, though levels for IL-5 and IL-13 did not reach statistical significance. Stimulation with DEP alone did not affect any of the cytokines measured here. However, addition of DEP to the pI:C stimulation (pI:C+DEP) tended to reduce the release of most of the cytokines as compared to pI:C alone, and this reached statistical significance for IL-4. The release of cytokines by NK cells is important for linking the innate and adaptive immune response via recruiting and activating other immune cells . NK cell-derived cytokines activate dendritic cells [36, 37] and help to shape T cell responses after NK cells have migrated to secondary lymphoid compartments . The suppression of cytokine production of NK cells by DEP could result in an overall reduced or modified immune response.
Maturation and activation of NK cells in vivo depends on a complex microenvironment which defines the activation status of NK cells via soluble mediators as well as direct receptor-ligand interactions [6, 39–42]. In vivo various cytokines with activation potential for NK cells [4, 43–46] may be released by other cells types (e.g. dendritic cells, macrophages, T cells, and epithelial cells) and can change the activity and functionality of NK cells. In our study, isolated peripheral blood NK cell responses in the context of a viral mimic were investigated, which is a potential limitation of our data. However, since the cytotoxicity potential of NK cells exposed to DEP in vitro was heavily impaired and corresponds with our findings in human volunteers exposed to DE in vivo
, the effects shown here may have relevance for respiratory NK cells in vivo. Another potential limitation of our study is the use of the viral mimetic pI:C, which represents a model of double-stranded RNA (dsRNA) formed during viral replication. dsRNA is generated during infections with positive single stranded RNA (ssRNA) viruses (such as rhinovirus) and to a lower extent during infection with negative ssRNA viruses (such as influenza virus and RSV) . While this is a potential limitation, it also allowed us to examine more general mechanisms by which DEP affects NK cell function and avoid factors introduced by active infection with a replicative virus.
In addition to respiratory virus infections, NK cells also play a role in several pulmonary bacterial and fungal infections (as reviewed by ). For example, NK cells are important for the clearance of Pseudomonas aeruginosa from the lung . Exposure of mice to DE decreased bacterial clearance in exacerbated pulmonary pathology associated with Pseudomonas aeruginosa infection . The authors from this study concluded that exposure to DE increases susceptibility to Pseudomonas aeruginosa infection, resulting in increased epithelial damage. Similarly, NK cells play important roles during Mycobacteria tuberculosis infections by inducing apoptosis of infected monocytes/macrophages (reviewed by ). Exposure of mice to DEP prior to infection with Bacillus Calmette-Guerrin, a mouse model of M. tuberculosis in mice, showed that DEP impaired clearance of the pathogen . These studies also demonstrated that the number of NK cells or ability to produce IFN-γ was not impaired by DEP, but attributed the reduced clearance to lower macrophage responsiveness to IFN-γ. Neither one of these studies examined whether exposure to DEP affected the ability of NK cell to directly kill the pathogen or induce cell-mediated cytotoxicity in infected cells. It is possible that reduced NK cell-mediated cytotoxicity after DEP exposure may have contributed to the effects of DEP on Pseudomonas aeruginosa or M. tuberculosis infections.