Secretory autophagy holds the key to lysozyme secretion during bacterial infection of the intestine

ABSTRACT

In 2013, Dr. Lora Hooper and colleagues described the induction of antibacterial macroautophagy/autophagy in intestinal epithelial cells as a cytoprotective host defense mechanism against invading Salmonella enterica serovar Typhimurium (S. Typhimurium). Canonical autophagy functions in a primarily degradative capacity to safeguard cells and ensure survival during stress conditions, including pathogen infection. In contrast, secretory autophagy has emerged as an alternative nondegradative mechanism for cellular trafficking and unconventional protein secretion. More recently, a study by Bel et al. from Dr. Hooper's lab describes how intestinal Paneth cells exploit the endoplasmic reticulum (ER) stress response to release antibacterial lysozyme through secretory autophagy in response to S. Typhimurium infection.

Abbreviations

3-MA

3-methyladenine

ATG

autophagy-related

BFA

brefeldin A

DC

dendritic cell

EM

electron microscopy

EIF2AK3/PERK

eukaryotic translation initiation factor 2 alpha kinase 3

EIF2S1/eIF2α

eukaryotic translation initiation factor 2, subunit 1 alpha

ER

endoplasmic reticulum

GI

gastrointestinal

IL22

interleukin 22

ILC3

type 3 innate lymphoid cells

MAP1LC3/LC3

microtubule-associated protein 1 light chain 3

MYD88

myeloid differentiation primary response gene 88

SQSTM1

sequestosome 1

T_H17

T helper 17

TUDCA

tauroursodeoxycholic acid

TLR

toll-like receptor

XBP1

X-box binding protein 1

Main text

Epithelial cells, such as those that line the gastrointestinal (GI) tract, provide a physical and immunological barrier to pathogen invasion. Despite this, enteric pathogens are adept at traversing host barriers to enter cells and initiate infection. S. Typhimurium is an invasive bacterial pathogen that causes gastroenteritis in affected individuals, typically by the consumption of contaminated food and water. The Centers for Disease Control estimates that nontyphoidal Salmonella infections account for 1.2 million illnesses and approximately 450 deaths annually in the United States [1]. Therefore, there is a need to more fully understand the host-pathogen interactions that mediate Salmonella entry and infection in relevant mammalian GI cell types.

A recent study by Bel and colleagues further investigates the impact of S. Typhimurium infection on autophagy induction in small intestinal Paneth cells [2]. Paneth cells are specialized secretory cells of the epithelial lineage located at the base of intestinal crypts of Lieberkühn (for further review see refs [3].^, [4]). S. Typhimurium invades and induces autophagy in intestinal enterocytes in a manner that is dependent on the epithelial cell-intrinsic innate immune adaptor MYD88 (myeloid differentiation primary response gene 88) [5]. Bel et al. observed that S. Typhimurium also enters mouse Paneth cells and induces autophagy, as indicated by an increase in both size and number of MAP1LC3/LC3 (microtubule-associated protein 1 light chain 3)-positive puncta, a marker of autophagy.

The authors further characterized the contents of the large LC3-positive structures by a combination of approaches including immunofluorescence and electron microscopy (EM) in addition to immunoprecipitation. Through ultrastructural and immunogold-labeling EM analysis, the large Paneth cell granules were shown to contain lysozyme. These lysozyme-positive granules are surrounded by a double membrane in S. Typhimurium-infected cells (but not uninfected cells), demonstrating that bacterial infection interferes with normal lysozyme secretion. Paneth cells abundantly secrete lysozyme, in addition to other antimicrobial proteins. Lysozyme readily destroys bacteria through both peptidoglycan hydrolysis and also by a noncanonical cationic mechanism (reviewed in ref. [6]). Using a biochemical method, Bel and coworkers also demonstrated that LC3 and lysozyme co-immunoprecipitate in a manner that is dependent on S. Typhimurium, further supporting the idea that infection causes the rerouting of lysozyme through an alternate secretory route involving autophagy. Additionally, by using immunofluorescence microscopy the authors found that the lysozyme- and LC3-positive vesicles do not fuse with lysosomes as indicated by the absence of colocalization with the lysosomal marker CTSD (cathepsin D) or the autophagy cargo receptor SQSTM1/p62 (sequestosome 1), implying the granules are not targeted for degradation by autophagy. The authors also detected the accumulation of LC3-positive granules at the apical surface of Paneth cells and the release of lysozyme into the intestinal lumen in infected mice. Moreover, the authors verified that lysozyme- and LC3-positive vesicles colocalize with RAB8A, a GTPase involved in post-Golgi sorting, exocytosis and secretory autophagy [7].

Based on these observations, Bel and colleagues hypothesized whether secretory autophagy, an alternative mechanism whereby cells utilize components of the autophagy machinery to transport cargo to the extracellular space, was involved in the delivery of lysozyme from Paneth cells to the gut lumen. To test whether secretory autophagy is involved in S. Typhimurium-dependent lysozyme release from Paneth cells, the authors applied pharmacological inhibitors affecting different cellular trafficking pathways to harvested mouse intestinal crypts; these inhibitors include brefeldin A (BFA, an inhibitor of ER-Golgi transport) and chloroquine (an inhibitor of lysosomal acidification). Neither of these inhibitors affects lysozyme secretion in infected cells; however, lysozyme granules accumulate in infected Paneth cells in the presence of the autophagy inhibitor 3-methyladenine (3-MA). When a bacterial killing assay is performed using secretions from BFA-treated cells, a significant proportion of S. Typhimurium are killed compared to treatment with the secretions from cells treated with 3-MA, demonstrating that secretory autophagy is necessary for host antibacterial defense.

The authors then examined mice with a mutation in ATG16L1 (autophagy-related 16-like 1 [S. cerevisiae]) wherein Thr300 is replaced with an alanine residue (ATG16L1^T300A). ATG16L1 is the mammalian homolog of yeast Atg16 [8], and is a component of the heterotrimeric ATG12–ATG5-ATG16L1 complex involved in LC3 conjugation during autophagy. Individuals with the ATG16L1^T300A single-nucleotide polymorphism have an increased predisposition to developing Crohn disease [9]. Crohn disease and ulcerative colitis are the 2 major chronic inflammatory bowel diseases in humans (reviewed in ref. [10]). Crohn disease is characterized by chronic inflammation, most commonly in the ileum of the small intestine. Earlier work indicated that mice with hypomorphic Atg16l1 expression are associated with decreased Paneth cell lysozyme granule secretion [11]. In mice, ATG16L1^T300A is also associated with decreased antibacterial autophagy [12]. Additional studies have demonstrated that defects in bacterial clearance associated with ATG16L1^T300A are not unique to S. Typhimurium [13,14]. Despite these previous observations, the association between ATG16L1 mutations, defects in lysozyme export and secretory autophagy had never before been directly linked to Crohn disease pathologies.

In the current study, Bel et al. found that S. Typhimurium-infected intestinal crypts from mice harboring the ATG16L1^T300A mutation have defects in lysozyme secretion and bacterial killing. The addition of 3-MA does not further inhibit either lysozyme export in wild-type crypts, or bacterial killing in ATG16L1^T300A crypts, establishing a specific role for autophagy rather than an off-target effect of the drug. No lysozyme- and LC3-positive structures are observed in crypts from ATG16L1^T300A mice. Taken together, these results provide support for a model in which secretory autophagy defects contribute to abnormal lysozyme trafficking and export into the intestinal lumen in Crohn patients with Atg16l1 mutations.

Next, the authors investigated the intracellular signals involved in stimulating secretory autophagy in S. Typhimurium-infected Paneth cells and found evidence for upregulation of the ER stress response. Invasive S. Typhimurium causes Golgi fragmentation, which does not occur when Paneth cells are infected with a non-invasive mutant (invA∆) or the commensal Bacteroides thetaiotaomicron (neither of which induce lysozyme secretion). However, the authors observed increased protein levels of DDIT3/CHOP (DNA-damage inducible transcript 3), indicating that infection by S. Typhimurium induces the ER stress response most likely as a result of disrupted ER-Golgi trafficking. Similarly, application of the ER stress-inducing agent thapsigargin or the ER stress inhibitor tauroursodeoxycholic acid (TUDCA) elevate or reduce DDIT3 levels and lysozyme secretion, respectively. Furthermore, the authors found that secretory autophagy release of lysozyme is mediated through the EIF2AK3/PERK (eukaryotic translation initiation factor 2 alpha kinase 3)-EIF2S1/eIF2α (eukaryotic translation initiation factor 2, subunit 1 alpha) arm of the ER stress pathway as indicated by increased phosphorylation of both EIF2AK3 and EIF2S1 in the intestines of S. Typhimurium-infected mice. Secretory autophagy of lysozyme is induced when uninfected mice are treated with salubrinal, an inhibitor of p-EIF2S1 dephosphorylation, further supporting the role of the ER stress response in promoting secretory autophagy-dependent export of lysozyme in Paneth cells. Antibacterial defense is compromised when secretory autophagy of lysozyme is inhibited upon treatment of infected mice with TUDCA.

Antibacterial autophagy in intestinal enterocytes requires MYD88 [5], which functions downstream of most innate immune toll-like receptors (TLRs). In Paneth cells, secretory autophagy of lysozyme is inhibited in S. Typhimurium-infected mice with a dendritic cell (DC) Myd88 defect; this is not the case for mice with an epithelial cell-specific deletion of Myd88. The authors conclude that DC MYD88 is necessary for secretory autophagy in Paneth cells. Next, Bel and coworkers explored the requirement for DC MYD88 and the possible involvement of type 3 innate lymphoid cells (ILC3) and secretion of IL22 (interleukin 22) for antigen presentation to epithelial cells by DC TLRs. Secretory autophagy is reduced in infected rorc^−/− mice, which are deficient in ILC3 and T helper 17 (T_H17) cells (but secretory autophagy is not affected in T cell-deficient rag1^−/− mice), supporting the role of ILC3 in mediating S. Typhimurium-dependent secretory autophagy in Paneth cells. Release of lysozyme via secretory autophagy is restored in infected myd88^−/− mice treated with recombinant IL22, with no subsequent impact on lysozyme transcript levels. All together, these studies point to the role of the DC-ILC3 pathway in activating secretory autophagy of antibacterial lysozyme in Paneth cells following S. Typhimurium infection and induction of the ER stress response.

ATG16L1^T300A is a major Crohn disease risk variant [15]. Crohn disease susceptibility has also been associated with transcription factor XBP1 (X-box binding protein 1), which is required for ER expansion during stress [16]. Additional genetic factors have been identified that are beyond the scope of this review (for example, see ref. [4,9,10,15]). The work presented by Bel et al. demonstrates how a combination of hereditary factors (such as single-nucleotide polymorphisms in ER stress or autophagy-related genes) and environmental triggers (invasive pathogen infection) can contribute to an inflammatory state and predisposition towards the risk of developing Crohn disease. Secretion of antimicrobials in the gut is essential for maintaining a stable population of microbiota. This study provides further evidence for how defects in lysozyme export (or other antimicrobial proteins secreted by Paneth cells) could lead to the overgrowth of harmful pathogenic bacteria and an altered symbiotic commensal population, contributing to the chronic inflammatory state characteristic of Crohn disease.

Therefore, it appears that autophagy has dual roles in maintaining appropriate control over harmful bacterial populations in the gut—both by classic xenophagy [5,17] and release of lysozyme via secretory autophagy in response to pathogen invasion [2]. As with most diseases, the underlying risk factors and pathophysiology are often unique to a particular individual. Additional understanding of how specific gene variants and their associated proteins function in the cell is critical to not only furthering our knowledge of the basic cell biology of these processes but essential as we move towards more personalized disease treatment approaches. Further work is necessary to elucidate the likely myriad of roles that secretory autophagy plays in mediating cell and tissue homeostasis.

Funding Statement

This work was supported by the National Institute of General Medical Sciences under grant [grant number GM053396].

Acknowledgments

The authors thank Drs. Lora Hooper and Shai Bel for helpful comments.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.