11) in the NOD1-deficient animals when compared to WT controls. These data suggest that NOD1 deficiency impairs recruitment of inflammatory
cells to the lung during Lp infection. We next measured levels of cytokines and chemokines to examine the mechanism of NOD1-mediated protection. Cytokine levels from lung homogenates from WT, Nod1−/−, and Nod2−/− animals were measured for TNFα, IL-1β, IL-6, KC, IL-18, and MCP-1 to determine if there were significant differences in WT compared to Nod1−/− and Nod2−/− animals (Fig. 5). At 4 h, there was significantly decreased production of IL-1β (WT 1.00±0.06 versus NOD1 0.68±0.06 (mean±SEM)), KC (WT 1.00±0.12 versus NOD1 0.72±0.05), and trend toward decreased TNFα (WT 1.00±0.07 versus NOD1 0.78±0.09, p=0.06) in the Nod1−/− animals, when compared to WT controls (Fig. 5A, C, and G). In contrast, at 4 h, there was no change in IL-6, see more IL-18, or MCP-1 levels in the Nod1−/− animals (Fig. 5B, H, and I). At 24 h, Nod1−/− animals exhibited significantly increased levels of IL-6 production (WT 1.00±0.06 versus NOD1 1.35±0.13) compared to WT controls and a trend toward increased TNFα production (WT 1.00±0.08 versus NOD1 1.36±0.19, p=0.06). The only significant change seen in the Nod2−/− animals compared to WT controls at 4 h was a significantly increased production of IL-6 ZD1839 (WT 1.00±0.36 versus NOD2 1.49±0.66) and
MCP-1 (WT 1.00±0.12 versus NOD2 2.04±0.49). In addition, significant increases were seen in Nod2−/− animals compared to WT in IL-1β (WT 1.00±0.19 versus NOD2 1.49±0.43), IL-6 production (WT 1.00±0.11 versus NOD2 from 1.49±0.20), and MCP-1 production (WT 1.00±0.10 versus NOD2 1.55±0.21) at 24 h (Fig. 5E, F, and L). In addition, IFN-γ was analyzed at the 24-h, 72-h and 10-day time points and only minimal production was seen in lung homogenates (our unpublished observations). The levels of IFN-γ were not different when comparing WT, Nod1−/−, and Nod2−/− mice. These data demonstrate
an early impaired production of proinflammatory cytokines KC and IL-1β seen in the absence of NOD1 protein and a later increase in proinflammatory markers (IL-1β, IL-6, and MCP-1) in Nod2−/− and (IL-6) Nod1−/− animals. Our data herein suggest that both NOD1 and NOD2 can detect Lp, but only NOD1 regulates in vivo bacterial clearance at 72 h. In addition, NOD1-deficient animals display early decreases in PMN recruitment to the alveolar space of the lung at 4 and 24 h and NOD2-deficient animals display a significant increase in PMN recruitment at 24 h. NOD1- and NOD2-deficient mice also show altered pulmonary inflammatory cell infiltration and cytokine responses to Legionella. In our aerosolized animal model, we identified higher Lp CFU in Nod1−/− mice compared to WT controls. Delayed bacterial clearance of Lp has been a characteristic of other knockout systems.