Contribution of the Purinergic Receptor P2X7 to Development of Lung Immunopathology during Influenza Virus Infection

ABSTRACT An exacerbated immune response is one of the main causes of influenza-induced lung damage during infection. The molecular mechanisms regulating the fate of the initial immune response to infection, either as a protective response or as detrimental immunopathology, are not well understood. The purinergic receptor P2X7 is an ionotropic nucleotide-gated ion channel receptor expressed on immune cells that has been implicated in induction and maintenance of excessive inflammation. Here, we analyze the role of this receptor in a mouse model of influenza virus infection using a receptor knockout (KO) mouse strain. Our results demonstrate that the absence of the P2X7 receptor results in a better outcome to influenza virus infection characterized by reduced weight loss and increased survival upon experimental influenza challenge compared to wild-type mice. This effect was not virus strain specific. Overall lung pathology and apoptosis were reduced in virus-infected KO mice. Production of proinflammatory cytokines and chemokines such as interleukin-10 (IL-10), gamma interferon (IFN-γ), and CC chemokine ligand 2 (CCL2) was also reduced in the lungs of the infected KO mice. Infiltration of neutrophils and depletion of CD11b+ macrophages, characteristic of severe influenza virus infection in mice, were lower in the KO animals. Together, these results demonstrate that activation of the P2X7 receptor is involved in the exacerbated immune response observed during influenza virus infection.

P2X7 receptor (Fig. 1A). One hour before infection with influenza A/Puerto Rico/8/1934 H1N1 (PR/8) virus or A/Netherlands/602/2009 H1N1 pandemic (H1N1pdm) (NL/09) virus, cells were treated with BBG, BzATP, or AZ11645373. No significant differences were observed in PR/8 virus growth in cells treated with BBG or BzATP compared to the untreated cells (P ϭ 0.1) (Fig. 1B). We observed a significant reduction in virus titers in cells pretreated with AZ and infected with PR/8 at 24 h and 48 h postinfection compared to untreated cells (P Ͻ 0.05) (Fig. 1B). In cells infected with NL/09 virus and treated with BBG or BzATP, no significant differences were observed in virus growth (Fig. 1C). In cells pretreated with AZ and infected with NL/09 virus, we observed a significant reduction in virus titers at 24 h postinfection compared to untreated cells (P Ͻ 0.05) (Fig. 1C).
Mice lacking P2X7 receptors have a better outcome after influenza virus infection. Activation of the P2X7r has been associated with an increased and sustained inflammatory response. To determine whether this pathway is related to the exacerbated lung immunopathology developed after influenza virus infection, we tested mice lacking the P2X7r in our mouse model of influenza virus infection. After infection with PR/8 virus, we observed a significant reduction in body weight at days 8 and 9 postinfection in the wild-type (WT) mice (P Ͻ 0.05) ( Fig. 2A). We also observed increased survival in the virus-infected P2X7r KO mice (40%) compared to the wild-type mice (0%) (P Ͻ 0.05) (Fig. 2B). Similarly, in wild-type mice infected with NL/09 virus, we P2X7 Receptor in Influenza Immunopathology ® observed a significant reduction in body weight at days 4 to 8 postinfection (P Ͻ 0.05) (Fig. 2C). A significant increase in survival was observed in the infected P2X7r KO mice (50%) compared to the infected wild-type mice (0%) (P Ͻ 0.05) (Fig. 2D).
Reduced histopathology in the lungs of P2X7r KO mice infected with the NL/09 virus. The amount of replicating virus in the lungs is thought to correlate with the degree of lung pathology. Therefore, we infected wild-type and P2X7r KO mice with NL/09 virus, and at 4 or 7 days after infection, the lungs were harvested, and viral titers were determined by plaque assay. Reduced virus titers were observed in the lungs of the P2X7r KO mice compared to the wild-type mice, particularly in the samples collected on day 7 postinfection, although these differences were not statistically significant (P ϭ 0.1) (Fig. 3A). In a different set of mice, the lungs were collected on day 4 postinfection, and representative sections were prepared for histopathology. In both groups of mice, lesions were characterized as bronchointerstitial pneumonia, although the severity of the lesions differed in the groups (P Ͻ 0.05) ( Table 1). In the lungs of the virus-infected P2X7r KO mice, the bronchiolar region showed mild to moderate epithelial degeneration, characterized by vacuolation and multifocal necrosis of individual epithelial cells with a thin layer of peribronchiolar inflammatory cells, and moderate perivascular inflammation with little to no presence of intraluminal debris (Fig. 3B). The bronchiolar region of the lungs from wild-type mice exhibited a marked epithelial degeneration characterized by vacuolation, segmental necrosis, and loss of the epithelium, with a thick layer of peribronchiolar inflammatory cells, and mild to marked amounts of intraluminal debris (Fig. 3C). At the alveolar level, lesions in the KO mice consisted of mild inflammation as well as tissue damage that was characterized by the presence of scattered cell necrosis with mild to moderate infiltration of inflammatory cells (Fig. 3D). The alveolar region of lungs from wild-type mice exhibited marked alveolar inflammation characterized by multifocal necrosis and marked infiltration by inflammatory cells (Fig. 3E).
Reduction of apoptosis in lungs of mice lacking the P2X7r and infected with influenza A virus. Signaling through the P2X7 receptor induces cell apoptosis by activation of caspase cascades, including caspase-1, -3, and -8. To determine the levels of apoptosis after infection with influenza A NL/09, we quantified the number of  Fig. 4A and B). After quantification, we confirmed that the number of cleaved caspase-3-positive cells was significantly lower in the lungs of the infected KO mice (P Ͻ 0.05), (Fig. 4C), indicating a reduced amount of apoptosis in the lungs of these mice.
The production of proinflammatory cytokines is lower in the virus-infected P2X7r KO mice than in the wild-type mice. To determine whether the reduced pathology and apoptosis observed in the lungs of the P2X7r KO mice were correlated with decreased cytokine production after infection, we first quantified the amount of cytokines in ex vivo-infected bone marrow-derived macrophages (BMDM) and then in the lungs of both virus-infected P2X7r KO and WT mice. Cultured BMDM, obtained from P2X7r KO and WT mice, were infected with NL/09 virus, and supernatants were collected at 24 and 48 h postinfection. We observed an increased production of CXC  a Lung pathology was scored on a scale of 0 to 4. For amount of tissue affected, lung pathology was scored as follows: 0, none; 1, Ͻ25% of section; 2, 25 to 50% of section; 3, 50 to 75% of section; 4, Ͼ75% of section. For inflammation and epithelial degeneration, lung pathology was scored as follows: 0, no lesions; 1, mild changes with scattered cell necrosis/vacuolation and few/scattered inflammatory cells in the bronchiolar epithelium with minimal perivascular inflammation; 2, moderate multifocal vacuolation and/or cell necrosis with peribronchiolar and/or perivascular inflammation (Ͻ5 cell layers thick); 3, marked, multifocal/segmental necrosis, epithelial loss/effacement and peribronchiolar and/or perivascular inflammation (Ͼ5 cell layers thick); 4, severe, coalescing areas of necrosis, parenchymal effacement with confluent areas of inflammation. For intraluminal debris, lung pathology was scored as follows: 0, none; 1, mild amounts; 2, moderate amounts; 3, marked amounts; 4, severe amounts. Each score value represents the mean of five mice per group. Overall pathology scores for the P2X7 KO group that are significantly different (P Ͻ 0.05) from those of the wild-type group are indicated by an asterisk. ( Fig. 6). A trend to significance was observed in the amount of IL-2 in the lungs of the KO mice compared to the lungs of the WT mice (P ϭ 0.08) (Fig. 6). Similar results were observed in mice infected with a different influenza virus, PR/8, with significantly larger amounts of IL-10, IFN-␥, CXCL10, CCL2, and CCL5 in the lungs of the WT mice (see Fig. S1 in the supplemental material). The percentages of pulmonary activated macrophages and activated lymphocytes were higher in the P2X7 KO mice after infection. We evaluated the myeloid cell population in the whole lungs of P2X7 KO and WT mice infected with NL/09 virus and collected on day 3 or day 7 postinfection (p.i.). We observed a higher number of CD11b ϩ macrophages in the lungs of virus-infected P2X7 KO mice at both time points compared to the virus-infected WT mice (P ϭ 0.056) ( Fig. 7A) with no significant differences in the absolute number of cells between mock-infected or virus-infected WT and P2X7 KO samples.
We then evaluated the leukocyte population in the bronchoalveolar lavage fluid (BALF) of both P2X7 KO and WT mice 3 and 7 days postinfection with NL/09 virus. The total number of cells and the viability in BALF were not different between the groups. The percentage of activated macrophages was significantly increased at day 3 p.i. in the P2X7 KO mice than in the WT mice (P Ͻ 0.05), but not at day 7 p.i. (Fig. 7B). The percentage of neutrophils in the BALF samples was significantly higher at day 3 p.i. in

P2X7 Receptor in Influenza Immunopathology
® the WT mice than in the P2X7 KO mice (Fig. 7B) (P Ͻ 0.05). No differences between groups were observed at day 7 p.i. A significantly higher percentage of activated lymphocytes was observed in the KO mice at day 7 p.i. (P Ͻ 0.05) (Fig. 7B). No differences in the percentage of activated lymphocytes were observed between groups at day 3 p.i. The percentage of lysed cells in the BALF of WT mice tended to be higher, especially at day 7 p.i. (P ϭ 0.1) (Fig. 7B).

DISCUSSION
The mechanisms by which the immune response is balanced to achieve influenza virus clearance and tissue repair versus harmful immunopathology are not well understood (5,6,33). In this study, we showed that mice lacking the purinergic receptor P2X7 have a better clinical outcome after influenza A virus infection characterized by an increased survival rate with an overall reduced immunopathology of the lungs compared to the wild-type mice.
ATP, the main ligand for P2X7 receptor, is usually located intracellularly, although small amounts can be found in the extracellular space (10,34,35). When cells are under stress or damaged by infection, ATP is released into the extracellular space and rapidly  binds to excitatory receptors such as P2X7 to promote an inflammatory response (9,10,34). Activation of the purinergic pathway through P2X7r encompasses a series of inflammation-related responses, including activation, proliferation, and recruitment of immune cells, as well as secretion of diverse inflammatory mediators, including cytokines and chemokines (7,9,10).
The initial evaluation was done in vitro by infecting A549 cells and treating them with different P2X7r agonist and antagonists before influenza virus infection to determine the impact on virus growth. We observed differences only in cells treated with the P2X7r antagonist AZ11645373. AZ is a potent cyclic imide (50% inhibitory concentration [IC 50 ], 5 to 10 nM) that selectively inhibits the human P2X7r (36). In contrast, BBG, the other P2X7r antagonist used in this study, is moderately selective and less potent in blocking the human P2X7r (IC 50 , 200 nM) (37). Since we treated cells with the same concentration of compound (100 M), we speculate that differences in potency may be one of the reasons why we observed some effect in virus growth with one receptor antagonist but not with the other.
For the in vivo studies, we first observed that mice lacking the P2X7 receptor (KO mice) had better survival after infection and that survival was independent of the virus strain used. It is generally accepted that survival of the infected host correlates with a reduction of the viral burden in the lung (38,39). However, when we quantified the virus titers in the lungs of infected mice, we observed that the virus titers in the lungs of virus-infected KO mice were not significantly different from those in the lungs of virus-infected wild-type mice. This lack of correlation between lung virus titers and severity of disease has been previously reported (6,40,41). In early stages of viral P2X7 Receptor in Influenza Immunopathology ® infection, influenza A virus (IAV) activates antiapoptotic signals to evade the host defense mechanism. However, in the later stages of infection, IAV triggers apoptosis through activation of intrinsic pathways in order to achieve enhanced viral replication and dissemination (42,43), contributing to tissue damage. We evaluated apoptosis in lungs at day 4 postinfection in what can be considered late in infection, and we observed reduced numbers of apoptotic cells in the KO mice. Apoptosis can occur through many mechanisms, including the production of proinflammatory cytokines leading to an increased burden of apoptosis of the alveolar epithelium (44), activation of caspase cascades, including caspase-1, -3, and -8 (45), and the formation and release of reactive oxygen species (46). It is interesting that activation of the P2X7 receptor induces some of the aforementioned proapoptotic mechanisms (45,46); therefore, the absence of this receptor is causing a reduction in apoptosis during later stages of infection.
The extent and severe damage observed in the lung tissues of the virus-infected WT mice can be due to both the cytolytic effect of the virus infection itself and to immunopathology caused by immune cells targeting infected cells. We did not find a significant difference in lung virus titers between the virus-infected KO and WT mice; therefore, the reduced histopathology and apoptosis observed in the KO mice pointed toward differences in immune-mediated pathology.
During influenza virus infection, an increased production of proinflammatory cytokines positively correlates with severity of disease in humans, especially during infection with highly virulent strains of the virus (38,47,48). During the recent H1N1 pandemic, patients with severe disease also had high levels of proinflammatory cytokines (39). A similar correlation has been observed in animal models of influenza virus infection (49,50). In the mouse model of influenza, mice with severe clinical signs that succumb to infection have increased levels of proinflammatory cytokines and chemokines in the lungs and develop lung immunopathology (50)(51)(52). In our studies, we observed that the virus-infected P2X7r KO mice had lower levels of proinflammatory cytokines such as IL-6, TNF-␣, IFN-␥, CXCL10, and CCL2 in their lungs at day 3 postinfection. Activation of P2X7 receptors by extracellular ATP plays a central role in the release of proinflammatory cytokines, and it plays a role on the regulation of macrophages, neutrophils, and effector T cells, important cell mediators during inflammation (9, 11, 53, 54). Therefore, the lower level of cytokines observed in the lungs of the P2X7r KO mice after influenza virus infection may be explained by the lack of activation of this purinergic signaling pathway in these mice. We observed that IL-10 levels in the lungs of the KO mice infected with PR/8 or NL/09 were lower than those of the WT mice. IL-10 is considered an anti-inflammatory cytokine in the context of influenza virus infection, and its production relates to reduced immunopathology and increased survival (41). This discrepancy may be explained by the time when we measured IL-10 in the lungs, since the main contributors to IL-10 production in the context of influenza infection are CD8 ϩ effector T cells, a population of cells that is found later in infection (41). In addition, increased levels of this cytokine early during infection can hinder the activation of the antiviral immune response (53), suggesting that the increased levels observed in the WT mice early during infection may be detrimental.
The suppression of cytokine production at the site of infection using drugs (55) or mice with specific cytokines knocked out (56-58) is not enough to confer full protection against lethality during influenza virus infection. The lack of the receptor in our model does not induce a global reduction in cytokine production after influenza virus infection since we did not observe differences in cytokine levels in serum from the virus-infected KO and WT mice. We also measured the basal levels of IL-1␤, IL-6, TNF-␣, and CCL5 in the lungs of naive KO and WT mice, and no significant differences were observed (see Fig. S2 in the supplemental material). Together, these results suggest that the reduced histopathology and increased survival observed in the KO mice after infection are in part due to a protective immune response at the site of infection rather than ablation of the response.
The local increase in cytokines stimulates the infiltration of neutrophils as well as monocytes from the periphery, an essential step for resolution of virus infection (2,59). In our studies, we observed a reduction in neutrophil influx to the airways of the KO mice at day 3 postinfection that was also evident during the analysis of the lung histology sections. An exacerbated response with an increased influx of highly proinflammatory neutrophils and the prolonged activation of these cells leads to increased tissue damage and lethality (3,6,60). Purinergic signaling in the presence of high concentrations of extracellular ATP has a critical role in activation of neutrophils for the production of cytokines and in the increased influx and persistence of neutrophils at the site of infection (11). Furthermore, we observed a significantly smaller amount of CXCL10 in the lungs of KO mice infected with NL/09 or PR/8 influenza virus. This chemokine is an important factor for the chemotaxis and activation (increased oxidative burst) of neutrophils, and high levels in the respiratory tract after influenza virus infection correlate with enhanced pulmonary immunopathology (61)(62)(63). Together, these results suggest that purinergic signaling through the P2X7 receptor is important for the recruitment and possible activation of neutrophils in the lungs after influenza virus infection.
We observed an increased influx of activated macrophages to the airway space in the KO mice compared to the WT group. In addition, flow cytometry analysis of the lungs in virus-infected mice showed an increased number of CD11b ϩ macrophages or inflammatory macrophages in the KO mice compared to the WT mice. These cells can be proinflammatory or immunomodulatory based on the cytokine profile they produce and surface markers they express after activation (1). Lung accumulation of these inflammatory macrophages with an immunoregulatory phenotype contributes to reduce the development of immunopathology (64). Furthermore, it has been shown that maintenance of this cell population for a longer time at the site of infection correlates with protection of mice from influenza-induced immunopathology and lethality (65). The role of the P2X7 receptor in the activation of macrophages has been previously established in vivo, ex vivo, and in vitro (8). Activation of the P2X7r pathway is important for the release of proinflammatory cytokines, increases in cation fluxes, respiratory burst, inhibition of phagocytosis, and apoptosis in different populations of macrophages (19,21,66).
Our results indicate that not only the number of immune cells infiltrating the lung after infection determines the development of an exacerbated response leading to immune pathology, but and perhaps more important, is the type and activation status of the cells recruited to the site of infection. The histopathology analysis of lung sections clearly showed an increased number of infiltrating cells in the lung parenchyma of the virus-infected wild-type mice compared to the virus-infected KO mice. However, further analysis of the infiltrating cells showed that the number of CD11b ϩ macrophages (whole lung) and number of activated macrophages (BALF), both associated with protection, were higher in the virus-infected KO mice. Furthermore, activation of the P2X7 receptor can modify the protein secretion profile of stimulated macrophages depending on the activation status of the cells (proinflammatory or immunomodulatory), providing evidence that activation of this pathway may function to fine tune the activity of the heterogeneous population of macrophages observed through infection of the lungs (54).
Activation of the purinergic receptor P2X7 signaling by increased levels of extracellular ATP caused by influenza virus infection leads to an exacerbated immune response characterized by increased production of proinflammatory cytokines, induction of apoptosis, increased influx of neutrophils in the airways, and development of lung histopathology. This response likely produced an amplified influenza virus-induced immunopathology, leading to an increased morbidity and mortality in the infected mice. On the contrary, these effects were moderated in the P2X7r KO mice, resulting in a better outcome to infection.
Our studies demonstrated an important role for this receptor in the development of lung immunopathology during influenza virus infection. These results contribute to a better understanding on how the immune response may be directed to cause the immunopathology observed during influenza infection. Furthermore, understanding the role this receptor has in the activation of the immune response opens an opportunity for development of medical interventions aimed to modulate the host response toward a balanced immune response that will clear the virus with minimum lung damage.

MATERIALS AND METHODS
Purinergic receptor agonists and antagonists. Adenocarcinomic human alveolar basal epithelial cells or A549 cells (ATCC, Manassas, VA) were evaluated to determine the presence of the purinergic receptor P2X7. A549 cells cultured on cover glass were fixed with 4% buffered formalin and later incubated with 5% nonfat milk for 30 min. After the blocking step, the cells were incubated with a polyclonal anti-P2X7 receptor antibody raised in rabbit (OriGene, Rockville, MD). This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals (67).
Female C57BL/6 or P2X7 receptor knockout mice (8 to 10 weeks old) purchased from Jackson Laboratories (Bar Harbor, ME) were used for all experiments. For virus challenges, mice were anesthetized by intraperitoneal injection of a mixture of ketamine (100 mg/kg of body weight) and xylazine (5 mg/kg) before intranasal administration of a lethal dose of influenza virus PR/8 or NL/09 in a volume of 30 l. Animals were monitored daily for clinical signs of illness, and body weights were recorded daily for 14 days. Upon reaching 75% of initial body weight, animals were humanely euthanized.
Lung virus titers. On days 4 and 7 postinfection, groups of mice were euthanized, and the lungs were collected and homogenized (BeadBlaster 24; Benchmark Scientific) in 1 ml of sterile phosphatebuffered saline (PBS). The lung homogenates were spun at 16,000 ϫ g for 10 min to pellet tissue debris, and the supernatants were collected. Samples were stored at Ϫ80°C until titration was performed by standard plaque assay on MDCK cells.
Histopathology. Mice were euthanized on day 4 or 7 postinfection, and their lungs were perfused in situ with 1 ml of 10% buffered formalin. Lung sections were stained with hematoxylin-eosin and evaluated by an experienced comparative pathologist from the Center of Comparative Medicine and Surgery in the Icahn School of Medicine at Mount Sinai. The following score system (on a scale of 0 to 4) was used to analyze the histopathology of the lung sections: 0, no lesions; 1, mild changes with scattered cell necrosis/vacuolation and few/scattered inflammatory cells in the bronchiolar epithelium with minimal perivascular inflammation; 2, moderate multifocal vacuolation with peribronchiolar inflammation (Ͻ5 cell layers thick); 3, marked, multifocal/segmental necrosis, epithelial loss/effacement, and peribronchiolar inflammation (Ͼ5 cell layers thick); and 4, severe, coalescing areas of necrosis, parenchymal effacement with confluent areas of inflammation. Overall pathology scores per group were used for the statistical analysis of the results.
Additional sections were processed for immunohistochemistry (IHC) and used to determine the amount of the apoptosis-related protein caspase-3. Ten-micron-thick deparaffinized sections were subjected to heat-induced epitope retrieval by boiling in 10 mM sodium citrate buffer (pH 6) for 10 min, followed by a 30 min cool-down. Cleaved caspase-3 antibody (Cell Signaling Technology, Inc., Beverly, MA) was diluted 1:1,000 in the following buffer: 0.1 M PBS, 0.3% (wt/vol) bovine serum albumin (BSA), 0.1% (wt/vol) sodium azide, 0.06% (wt/vol) n-ethyl-maleimide, and 20% (vol/vol) glycerol. This antibody specifically recognizes the large fragment (17/19 kDa) of the active protein (cleaved), but not full-length caspase-3. Cleaved caspase-3-positive cells were counted in printed digital images using a transparent template with a rectangular box measured 0.25 mm by 0.5 mm for photographs taken at a magnification of ϫ20 in three different fields. An individual blind to the experimental treatments completed the quantification.
Bone marrow-derived macrophages. Bone marrow-derived macrophages (BMDM) were generated by culturing total bone marrow from femurs and tibias of P2X7 receptor KO or wild-type mice as previously described (68). Collected cells were plated at a density of 1 ϫ 10 6 cells/well in six-well plates using high-glucose Dulbecco modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 10 mM penicillin-streptomycin, 10 mM sodium pyruvate, 10 mM HEPES, 50 M 2-mercaptoethanol, and 20% L929 cell conditioned supernatant containing macrophage colony-stimulating factor (M-CSF). The cells were cultured for 5 days, after which the medium and nonadherent cells were removed, and fresh medium was added every other day until harvest between days 8 and 10. The cells were purified by adherence. To harvest the cells, we used 500 l of 0.5% trypsin. For infection studies, cells were plated in 12-well plates and infected with influenza A NL/09 virus at a multiplicity of infection (MOI) of 1 for 48 h. Supernatants of samples were collected at 24 and 48 h postinfection and analyzed for virus titer and cytokine production.
Cytokine and chemokine quantification. Protein levels of cytokines/chemokines were evaluated using a multiplex bead array assay (Milliplex; EMD Millipore, Billerica, MA) in cell culture supernatant of bone marrow-derived macrophages infected with influenza NL/09 or in lung homogenates and serum samples from mice infected with influenza virus NL/09 or PR/8 according to the manufacturer's instructions. The assays were run using 1,500 beads per set of each of cytokines measured per well in a total volume of 50 l. The plates were read on a Luminex MAGPIX platform. For each bead set, Ͼ50 beads were collected. The median fluorescence intensity of these beads was recorded and used for analysis with the Milliplex software using a 5P regression algorithm.
Bronchoalveolar lavage cytospin. Mice were euthanized by intraperitoneal administration of pentobarbital at the indicated time points. Leukocytes recruited to the airway space were collected by lavaging the lungs in situ by slowly injecting 1 ml of cold sterile PBS using an input 3-ml syringe with three-way stopcock and then approximately 1 ml of bronchoalveolar lavage fluid (BALF) was collected from lungs using the output 3-ml syringe. We repeated this procedure three times. Cell number and cell viability were determined using trypan blue and an automatic cell counter (Countess II; Thermo Fisher Scientific, Waltham, MA). For the cytospin, 300 l of each BALF sample containing 1 ϫ 10 5 cells was spun at 600 rpm for 10 min. The slides were air dried for 2 h and then stained with Diff-Quick. The slides were evaluated in a blind manner by a trained comparative pathologist. Differential cell counts (percentage) were established by counting at least 200 cells at a high magnification (ϫ600) using the following categorization (69). Undifferentiated alveolar macrophages were defined as relatively large, round to oval cells with eccentric round to oval nuclei and abundant granular cytoplasm, while activated macrophages were larger and contain more cytoplasm and numerous cytoplasmic vacuoles, with some macrophages binucleated. Neutrophils (10 to 15 m in diameter) were characterized by an eosinophilic cytoplasm and a basophilic lobulated nucleus with violet granules. Inactivated lymphocytes were small (Ͻ10 m) cells with rounded nuclei, dense nuclear chromatin, and little cytoplasm, while activated (reactive) lymphocytes were larger with more basophilic cytoplasm and eccentric nuclei. Finally, we also quantified the percentage of lysed cells by the relative proportion of isolated (naked) nuclei.
Statistical analysis. Statistical differences in virus titers and cytokine production between the KO and WT mice were analyzed using Student's t test analysis followed by the Holm-Sidak test. For analysis of survival rates, Kaplan-Meier curves were compared using the log rank (Mantel-Cox) test. For analysis of the flow cytometry results, we used a two-way analysis of variance followed by multiple comparisons using the Tukey posttest. Error bars indicate standard errors of the means. P values of Ͻ0.05 were considered significant. All the analyses were performed using the GraphPad software (La Jolla, CA).