Dr Louise Tzung-Harn Hsieh

Buruli ulcer (BU), caused by Mycobacterium ulcerans, is a devastating necrotizing skin disease. Key to its pathogenesis is mycolactone, the exotoxin virulence factor that is both immunosuppressive and cytotoxic. The discovery that the essential Sec61 translocon is the major cellular target of mycolactone explains much of the disease pathology, including the immune blockade. Sec61 inhibition leads to a loss in production of nearly all cytokines from monocytes, macrophages, dendritic cells and T cells, as well as antigen presentation pathway proteins and costimulatory molecules. However, there has long been evidence that the immune system is not completely incapable of responding to M. ulcerans infection. In particular, IL-1 beta was recently shown to be present in BU lesions, and to be induced from M. ulcerans-exposed macrophages in a mycolactone-dependent manner. This has important implications for our understanding of BU, showing that mycolactone can act as the "second signal " for IL-1 beta production without inhibiting the pathways of unconventional secretion it uses for cellular release. In this Perspective article, we validate and discuss this recent advance, which is entirely in-line with our understanding of mycolactone's inhibition of the Sec61 translocon. However, we also show that the IL-1 receptor, which uses the conventional secretory pathway, is sensitive to mycolactone blockade at Sec61. Hence, a more complete understanding of the mechanisms regulating IL-1 beta function in skin tissue, including the transient intra-macrophage stage of M. ulcerans infection, is urgently needed to uncover the double-edged sword of IL-1 beta in BU pathogenesis, treatment and wound healing.

The drivers of tissue necrosis in Mycobacterium ulcerans infection (Buruli ulcer disease) have historically been ascribed solely to the directly cytotoxic action of the diffusible exotoxin, mycolactone. However, its role in the clinically evident vascular component of disease aetiology remains poorly explained. We have now dissected mycolactone's effects on human primary vascular endothelial cells in vitro. We show that mycolactone-induced changes in endothelial morphology, adhesion, migration, and permeability are dependent on its action at the Sec61 translocon. Unbiased quantitative proteomics identified a profound effect on proteoglycans, driven by rapid loss of type II transmembrane proteins of the Golgi, including enzymes required for glycosaminoglycan (GAG) synthesis, combined with a reduction in the core proteins themselves. Loss of the glycocalyx is likely to be of particular mechanistic importance, since knockdown of galactosyltransferase II (beta-1,3-galactotransferase 6; B3GALT6), the GAG linker-building enzyme, phenocopied the permeability and phenotypic changes induced by mycolactone. Additionally, mycolactone depleted many secreted basement membrane components and microvascular basement membranes were disrupted in vivo during M. ulcerans infection in the mouse model. Remarkably, exogenous addition of laminin-511 reduced endothelial cell rounding, restored cell attachment and reversed the defective migration caused by mycolactone. Hence supplementing mycolactone-depleted extracellular matrix may be a future therapeutic avenue, to improve wound healing rates.

Buruli ulcer (BU) is a neglected tropical disease caused by subcutaneous infection with Mycobacterium ulcerans and its exotoxin mycolactone. BU displays coagulative necrosis and widespread fibrin deposition in affected skin tissues. Despite this, the role of the vasculature in BU pathogenesis remains almost completely unexplored. We hypothesise that fibrin-driven ischemia can be an 'indirect' route to mycolactone-dependent tissue necrosis by a mechanism involving vascular dysfunction. Here, we tracked >900 vessels within contiguous tissue sections from eight BU patient biopsies. Our aim was to evaluate their vascular and coagulation biomarker phenotype and explore potential links to fibrin deposition. We also integrated this with our understanding of mycolactone's mechanism of action at Sec61 and its impact on proteins involved in maintaining normal vascular function. Our findings showed that endothelial cell dysfunction is common in skin tissue adjacent to necrotic regions. There was little evidence of primary haemostasis, perhaps due to mycolactone-dependent depletion of endothelial von Willebrand factor. Instead, fibrin staining appeared to be linked to the extrinsic pathway activator, tissue factor (TF). There was significantly greater than expected fibrin staining around vessels that had TF staining within the stroma, and this correlated with the distance it extended from the vessel basement membrane. TF-induced fibrin deposition in these locations would require plasma proteins outside of vessels, therefore we investigated whether mycolactone could increase vascular permeability in vitro. This was indeed the case, and leakage was further exacerbated by IL-1β. Mycolactone caused the loss of endothelial adherens and tight junctions by the depletion of VE-cadherin, TIE-1, TIE-2 and JAM-C; all Sec61-dependent proteins. Taken together, our findings suggest that both vascular and lymphatic vessels in BU lesions become "leaky" during infection, due to the unique action of mycolactone, allowing TF-containing structures and plasma proteins into skin tissue, ultimately leading to local coagulopathy and tissue ischemia.

The drivers of tissue necrosis in Mycobacterium ulcerans infection (Buruli ulcer disease) have historically been ascribed solely to the directly cytotoxic action of the diffusible exotoxin, mycolactone. However, its role in the clinically-evident vascular component of disease aetiology remains poorly explained. We have now dissected mycolactone's effects on primary vascular endothelial cells in vitro and in vivo. We show that mycolactone-induced changes in endothelial morphology, adhesion, migration, and permeability are dependent on its action at the Sec61 translocon. Unbiased quantitative proteomics identified a profound effect on proteoglycans, driven by rapid loss of type II transmembrane proteins of the Golgi, including enzymes required for glycosaminoglycan (GAG) synthesis, combined with a reduction in the core proteins themselves. Loss of the glycocalyx is likely to be of particular mechanistic importance, since knockdown of galactosyltransferase II (beta-1,3-galactotransferase 6; B3Galt6), the GAG linker-building enzyme, phenocopied the permeability and phenotypic changes induced by mycolactone. Additionally, mycolactone depleted many secreted basement membrane components and microvascular basement membranes were disrupted in vivo. Remarkably, exogenous addition of laminin-511 reduced endothelial cell rounding, restored cell attachment and reversed the defective migration caused by mycolactone. Hence supplementing mycolactone-depleted extracellular matrix may be a future therapeutic avenue, to improve wound healing rates. Competing Interest Statement The authors have declared no competing interest.

The Mycobacterium ulcerans exotoxin, mycolactone, is responsible for the immunosuppression and tissue necrosis that characterizes Buruli ulcer. Mycolactone inhibits SEC61-dependent co-translational translocation of proteins into the endoplasmic reticulum and the resultant cytosolic translation triggers degradation of mislocalized proteins by the ubiquitin-proteasome system. Inhibition of SEC61 by mycolactone also activates multiple EIF2S1/eIF2α kinases in the integrated stress response (ISR). Here we show mycolactone increased canonical markers of selective macroautophagy/autophagy LC3B-II, ubiquitin and SQSTM1/p62 in diverse disease-relevant primary cells and cell lines. Increased formation of puncta positive for the early autophagy markers WIPI2, RB1CC1/FIP200 and ATG16L1 indicates increased initiation of autophagy. The mycolactone response was SEC61A1-dependent and involved a pathway that required RB1CC1 but not ULK. Deletion of Sqstm1 reduced cell survival in the presence of mycolactone, suggesting this response protects against the increased cytosolic protein burden caused by the toxin. However, reconstitution of baseline SQSTM1 expression in cells lacking all autophagy receptor proteins could not rescue viability. Translational regulation by EIF2S1 in the ISR plays a key role in the autophagic response to mycolactone. Mycolactone-dependent induction of SQSTM1 was reduced in eif2ak3−/-/perk−/- cells while the p-EIF2S1 antagonist ISRIB reversed the upregulation of SQSTM1 and reduced RB1CC1, WIPI2 and LC3B puncta formation. Increased SQSTM1 staining could be seen in Buruli ulcer patient skin biopsy samples, reinforcing genetic data that suggests autophagy is relevant to disease pathology. Since selective autophagy and the ISR are both implicated in neurodegeneration, cancer and inflammation, the pathway uncovered here may have a broad relevance to human disease.