To understand the pathogenesis of chronic inflammatory disease, we analyzed an experimental mouse model of a chronic lung disease that resembles asthma and chronic obstructive pulmonary disease (COPD) in humans. In this model, chronic lung disease develops after infection with a common type of respiratory virus is cleared to trace levels of noninfectious virus. Unexpectedly, the chronic inflammatory disease arises independently of an adaptive immune response and is driven by IL-13 produced by macrophages stimulated by CD1d-dependent TCR-invariant NKT cells. This innate immune axis is also activated in the lungs of humans with chronic airway disease due to asthma or COPD. These findings provide new insight into the pathogenesis of chronic inflammatory disease with the discovery that the transition from respiratory viral infection into chronic lung disease requires persistent activation of a novel NKT cell-macrophage innate immune axis. It has been widely speculated that infections are linked to the development of chronic inflammatory diseases. Although the connection between infection and chronic disease is uncertain, it likely depends on an aberrant immune response. In particular, it is believed that the innate immune system mediates the acute response to an infectious agent 1 , while an atypical adaptive immune response may cause chronic inflammation 2. Furthermore, infectioninduced alterations in the adaptive immune response that produce T cell or antibody-mediated

Increased mucus production is a common cause of morbidity and mortality in inflammatory airway diseases, including asthma, chronic obstructive pulmonary disease (COPD), and cystic fibrosis. However, the precise molecular mechanisms for pathogenic mucus production are largely undetermined. Accordingly, there are no specific and effective anti-mucus therapeutics. Here, we define a signaling pathway from chloride channel calcium-activated 1 (CLCA1) to MAPK13 that is responsible for IL-13-driven mucus production in human airway epithelial cells. The same pathway was also highly activated in the lungs of humans with excess mucus production due to COPD. We further validated the pathway by using structure-based drug design to develop a series of novel MAPK13 inhibitors with nanomolar potency that effectively reduced mucus production in human airway epithelial cells. These results uncover and validate a new pathway for regulating mucus production as well as a corresponding therapeutic approach to mucus overproduction in inflammatory airway diseases.

Chronic obstructive lung disease is characterized by persistent abnormalities in epithelial and immune cell function that are driven, at least in part, by infection. Analysis of parainfluenza virus infection in mice revealed an unexpected role for innate immune cells in IL-13-dependent chronic lung disease, but the upstream driver for the immune axis in this model and in humans with similar disease was undefined. We demonstrate here that lung levels of IL-33 are selectively increased in postviral mice with chronic obstructive lung disease and in humans with very severe chronic obstructive pulmonary disease (COPD). In the mouse model, IL-33/IL-33 receptor signaling was required for Il13 and mucin gene expression, and Il33 gene expression was localized to a virus-induced subset of airway serous cells and a constitutive subset of alveolar type 2 cells that are both linked conventionally to progenitor function. In humans with COPD, IL33 gene expression was also associated with IL13 and mucin gene expression, and IL33 induction was traceable to a subset of airway basal cells with increased capacities for pluripotency and ATP-regulated release of IL-33. Together, these findings provide a paradigm for the role of the innate immune system in chronic disease based on the influence of long-term epithelial progenitor cells programmed for excess IL-33 production.

Cytokine modulation of autophagy is increasingly recognized in disease pathogenesis, and current concepts suggest that type 1 cytokines activate autophagy, whereas type 2 cytokines are inhibitory. However, this paradigm derives primarily from studies of immune cells and is poorly characterized in tissue cells, including sentinel epithelial cells that regulate the immune response. In particular, the type 2 cytokine IL13 (interleukin 13) drives the formation of airway goblet cells that secrete excess mucus as a characteristic feature of airway disease, but whether this process is influenced by autophagy was undefined. Here we use a mouse model of airway disease in which IL33 (interleukin 33) stimulation leads to IL13-dependent formation of airway goblet cells as tracked by levels of mucin MUC5AC (mucin 5AC, oligomeric mucus/gel forming), and we show that these cells manifest a block in mucus secretion in autophagy gene Atg16l1-deficient mice compared to wild-type control mice. Similarly, primary-culture human tracheal epithelial cells treated with IL13 to stimulate mucus formation also exhibit a block in MUC5AC secretion in cells depleted of autophagy gene ATG5 (autophagy-related 5) or ATG14 (autophagyrelated 14) compared to nondepleted control cells. Our findings indicate that autophagy is essential for airway mucus secretion in a type 2, IL13-dependent immune disease process and thereby provide a novel therapeutic strategy for attenuating airway obstruction in hypersecretory inflammatory diseases such as asthma, chronic obstructive pulmonary disease, and cystic fibrosis lung disease. Taken together, these observations suggest that the regulation of autophagy by Th2 cytokines is cell-context dependent.

Some expressions and notations related to Equations 1 and 2 were presented incorrectly. The correct text and equations are below.The coefficients (β) in Cox's regression model are estimated by maximizing the partial likelihood function subject to a constraint on the L1-norm of the coefficients. The lasso estimator (β ) maximizes the objective function given below:Here l(β) is the log partial likelihood in the Cox model; for the exact form of this function, see ref. 41. The tuning parameter, λ in Equation 1, was chosen by 10-fold cross-validation. For the implementation, we used the R package "glmnet" (39).PROVAR was defined for each of the 222 TCGA samples as the sum of the estimated coefficients multiplied by protein expression levels, as shown below. Here i represents patients (i = 1,...,222), j represents proteins with nonzero coefficients (j = 1, ..., m), β j is the lasso coefficient of the jth protein marker, and X ij is the expression level of the jth protein for the ith patient. (Equation 2)The JCI regrets the error.