Abstract
Autophagy defects resulting in inflammation appear to be a key feature in the pathogenesis of Crohn colitis. An inflammatory colitis indistinguishable from Crohn disease is described in patients with chronic granulomatous disease (CGD). Patients with CGD have a mutated NADPH complex and are therefore deficient in reactive oxygen species (ROS) production; however, the underlying mechanism for the inflammatory colitis in CGD remained unknown. In a recent study, our group reported that NADPH-dependent ROS deficiency results in autophagic dysfunction that subsequently contributes to increased IL1B/interleukin 1β production. Mice deficient in the NADPH-complex component NCF4/p40phox, and CGD patients with a defect in NCF4 display minimal recruitment of LC3 to phagosomes in response to internalized bacteria and fungi. Human monocytes from patients with CGD with defective LC3 recruitment show increased IL1B production after LPS stimulation. Blocking IL1 protects NCF4-deficient mice from experimental colitis; importantly, improved clinical outcome in 2 CGD patients with colitis is also observed with IL1 blockade. Moreover, blocking IL1 restores defective autophagy in CGD mice and cells from patients with CGD. Thus, autophagic dysfunction underlies the pathogenesis of granulomatous colitis in CGD, and blocking IL1 can be used to treat CGD colitis.
Chronic granulomatous disease is a severe primary immunodeficiency formerly called “fatal granulomatous disease of childhood” because patients rarely survived childhood due to infections. CGD is caused by a mutation in one of the components of the NADPH oxidase complex resulting in the inability to produce NADPH oxidase-dependent reactive oxygen species (ROS), which are needed for phagocytic killing of microorganisms. This defect results in life-threatening infections especially due to Staphylococcus aureus and Aspergillus fumigatus, but despite the use of antimicrobial prophylaxis, these infections are still a major concern for patients with CGD. As an apparent paradox of ROS deficiency, patients with CGD exhibit a hyperinflammatory state that leads to the formation of granulomas and to an inflammatory colitis indistinguishable from that observed in Crohn disease. The resemblance of the inflammatory granulomatous colitis observed in Crohn disease and CGD suggests a shared underlying pathophysiology. This is further supported by the finding that polymorphisms in the region encoding NCF4 have been associated with Crohn disease in a GWAS study and that mutations in the NCF4 gene have recently been described as a cause of chronic granulomatous disease.
When we studied autophagy in NCF4−/− peritoneal macrophages and monocytes isolated from patients with CGD, we observed a profound defect in LC3-recruitment to the phagosome induced by bacteria. These data are consistent with previous reports that describe a role for ROS in the induction of autophagy. Similar to mice that are deficient in autophagy (Atg16l1 knockout), mice deficient in NADPH-oxidase as well as human cells from CGD patients secrete more IL1B in response to LPS. Blocking autophagy in human CGD monocytes does not result in augmented IL1B production, as observed in cells with intact NADPH oxidase. Thus, autophagy-dependent inhibition of IL1B is impaired in ROS-deficient cells. This supports the concept that autophagy defects in CGD resulting in IL1-mediated inflammation is a key feature in the pathogenesis of CGD. Testing this hypothesis, we were able to protect mice with TNBS-induced colitis by blocking IL1 with the receptor antagonist anakinra. Moreover, we treated 2 patients with CGD that had active colitis, including a patient with perirectal abscesses. Over the course of 3 mo, both patients responded with clinical improvement of the colitis and the perirectal abscesses healed.
These findings have important clinical consequences for the management of patients with CGD. Colitis is a major complication of CGD and is often treated with high doses of corticosteroids, which when given in childhood can result in severe growth retardation. Alternative therapies have been explored, such as anti-TNF, but due to increased risk of infection in CGD, blocking TNF results in fatalities. No infectious complications were seen during treatment with anakinra in our 2 patients despite a history of life-threatening infection. Therefore, anakinra might be a safe drug with efficacy equivalent or superior to corticosteroids without the severe consequences associated with long-term corticosteroid use.
Notably, the defects in autophagy that we describe in CGD are defects in a type of noncanonical autophagy called LC3-associated phagocytosis (LAP). We showed that anakinra is able to restore Aspergillus-induced LAP in murine and also in human CGD cells. Invasive aspergillosis is a major complication of CGD with a high mortality rate. Since LAP is a critical regulator of innate host defense, we tested the effect of blocking IL1 in a murine model of invasive pulmonary aspergillosis. Anakinra protects CGD mice from invasive pulmonary aspergillosis. The benefit of anakinra could be, in part, attributed to the reduction in pulmonary inflammation that is detrimental in CGD. However, fungal burden decreases significantly when CGD mice are treated with anakinra, indicating that restoring LAP in vivo during infection in CGD contributes to fungal clearance. Therefore, anakinra could be beneficial in life-threatening infections in CGD. In CGD patients with multiple liver abscesses due to S. aureus, the use of corticosteroids with antibiotic treatment is an accepted treatment regimen. We propose that anakinra therapy is more effective than corticosteroids. On the one hand, anakinra will dampen the IL1-driven detrimental inflammation during infection; on the other hand, it will restore defective LAP-mediated microbial clearing. Identifying defective autophagy due to deficient NADPH-oxidase ROS production unravels the paradox in CGD. Therefore, we demonstrate that CGD with deficient ROS results in defective autophagy and more inflammation, which is not dissimilar to IL1-driven inflammation of autoinflammatory diseases. Blocking IL1 will be beneficial in CGD by restoring deficient autophagy and decreasing detrimental IL1-mediated inflammation. Beyond the beneficial effects that anakinra has on CGD, we speculate that anakinra will also be beneficial in the setting of detrimental inflammation in diseases with defective autophagy, such as cystic fibrosis and Alzheimer disease.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.