The pathogenesis of chronic inflammatory diseases such as Crohn's disease (CD) is rooted in a combinatorial effect of host genetics and environmental triggers. We have previously found that CD patients with genetic mutations in ATG16L1 were prone to develop defective Paneth cell, and an abnormal Paneth cell phenotype is associated with poor clinical outcome. The morphological defect of small intestinal Paneth cell can therefore be used as a readout for such effect and sub-classify patients. Cigarette smoking is an important environmental affecter of CD and is associated with aggressive disease course in CD but the mechanism is unknown. We hypothesized that cigarette smoking affects Paneth cell function through interplay with host genetics.Methods:
We retrospectively analyzed all consecutive CD resection cases performed in our institution for 3 years (2011–2013). We collected the following clinical information: demographics, age at onset, age at operation, clinical phenotype (Paris classification), smoking and treatment history. Paneth cell phenotype was determined by lysozyme immunofluorescence. The clinical end point was time to recurrence after resection. Recurrence was defined by endoscopy (Rutgeert's score ≥ i2). For in vivo experiments, mice (with proper littermate controls) were subject to main-stream cigarette smoking using a smoking chamber (average 4 cigarettes/day; 5 days/week for 4 weeks). Crypt apoptosis was determined by cleaved caspase-3 immunohistochemistry. Microbiota composition was determined by 16s rRNA sequencing. Multivariate analysis was performed for statistical analysis and P value of 0.05 was considered significant.Results:
We found that CD patients with ATG16L1 T300A genotype who were smokers had a greater severity of Paneth cell defects as compared to patients with wild type ATG16L1 (P = 0.0488). This finding correlated with increased crypt apoptosis (P < 0.0001). Clinically, CD patients who were smokers and had abnormal Paneth cell phenotype had earlier recurrence after resection (P = 0.0081). When exposed to cigarette smoking, mice with Atg16l1 T300A developed defective Paneth cells and wild type littermates did not (P < 0.0001). This finding correlated with increased crypt apoptosis (P = 0.0030), indicating a direct functional role of Atg16l1 in Paneth cells may be required to mediate this effect. Exposure of Atg16l1-defensin-Cre (Paneth cell-specific deletion of Atg16l1) mice to cigarette smoking was sufficient to trigger abnormal Paneth cells. The process was associated with changes in microbiota composition, with Atg16l1 T300A mice showing increased abundance of certain invasive microbes (Prevotella) and reduced abundance of barrier microbes (Ruminococcus, Lactobacillus) after smoking compared with littermates. Cigarette smoking-induced abnormal Paneth cells in Atg16l1 T300A mice was also prevented by concomitant broad-spectrum antibiotics administration (P < 0.0001). TNF-α is a cytokine mediator for smoking-induced Paneth cell defects, as both pan-caspase inhibitor and anti-TNF-α were able to reverse the effect (P < 0.01 for both). Finally, macrophage of the reticulo-endothelial system (liver/spleen) were a potential source of TNF-α, as depletion of these cells with clodronate liposomes also prevented Paneth cell defects in Atg16l1 T300A mice (P = 0.0001).Conclusions:
We provided the first in vivo link that explains how cigarette smoking can act as a CD-associated environmental factor may lead to more aggressive disease through effects on Paneth cell function in a specific genetic background.