Mycobacterium tuberculosis (Mtb) is the causative agent of tuberculosis, an infectious disease with one of the highest morbidity and mortality rates worldwide. Leveraging its highly evolved repertoire of non-protein and protein virulence factors, Mtb invades through the airway, subverts host immunity, establishes its survival niche, and ultimately escapes in the setting of active disease to initiate another round of infection in a naïve host. In this review, we will provide a concise synopsis of the infectious life cycle of Mtb and its clinical and epidemiologic significance. We will also take stock of its virulence factors and pathogenic mechanisms that modulate host immunity and facilitate its spread. Developing a greater understanding of the interface between Mtb virulence factors and host defenses will enable progress towards improved vaccines and therapeutics to prevent and treat tuberculosis.

The heat shock protein 90 (Hsp90) is thought to be an excellent drug target against parasitic diseases. The leishmanicidal effect of an Hsp90 inhibitor, 17-N-allylamino-17-demethoxygeldanamycin (17-AAG), was previously demonstrated in both in vitro and in vivo models of cutaneous leishmaniasis. Parasite death was shown to occur in association with severe ultrastructural alterations in Leishmania, suggestive of autophagic activation. We hypothesized that 17-AAG treatment results in the abnormal activation of the autophagic pathway, leading to parasite death. To elucidate this process, experiments were performed using transgenic parasites with GFP-ATG8-labelled autophagosomes. Mutant parasites treated with 17-AAG exhibited autophagosomes that did not entrap cargo, such as glycosomes, or fuse with lysosomes. ATG5-knockout (Δatg5) parasites, which are incapable of forming autophagosomes, demonstrated lower sensitivity to 17-AAG-induced cell death when compared to wild-type (WT) Leishmania, further supporting the role of autophagy in 17-AAG-induced cell death. In addition, Hsp90 inhibition resulted in greater accumulation of ubiquitylated proteins in both WT- and Δatg5-treated parasites compared to controls, in the absence of proteasome overload. In conjunction with previously described ultrastructural alterations, herein we present evidence that treatment with 17-AAG causes abnormal activation of the autophagic pathway, resulting in the formation of immature autophagosomes and, consequently, incidental parasite death.

Lung disease due to non-tuberculous mycobacteria (NTM) is rising in incidence. Although both two-dimensional cell culture and animal models exist for NTM infections, a major knowledge gap is the early responses of human alveolar and innate immune cells to NTM within the human alveolar microenvironment. Here, we describe the development of a humanized, three-dimensional, alveolus lung-on-a-chip (ALoC) model of Mycobacterium fortuitum lung infection that incorporates only primary human cells, such as pulmonary vascular endothelial cells, in a vascular channel, and type I and II alveolar cells and monocyte-derived macrophages in an alveolar channel along an air–liquid interface. M. fortuitum introduced into the alveolar channel primarily infected macrophages, with rare bacteria inside alveolar cells. Bulk RNA sequencing of infected chips revealed marked upregulation of transcripts for cytokines, chemokines and secreted protease inhibitors (SERPINs). Our results demonstrate how a humanized ALoC system can identify critical early immune and epithelial responses to M. fortuitum infection. We envision potential application of the ALoC to other NTM and in studies of new antibiotics.

Macroautophagy/autophagy enables macrophages to degrade intracellular Mycobacterium tuberculosis (Mtb), and this defense depends on E3 ubiquitin ligases such as PRKN/PARKIN/PARK2 and SMURF1, which tag Mtb-associated structures for lysosomal clearance. Deubiquitinases (DUBs) counter ubiquitin ligases by removing ubiquitin from molecular targets. We hypothesized that DUBs might offset ubiquitin ligase activity and negatively regulate host immunity to Mtb. Here, we identify USP15 (ubiquitin specific peptidase 15) as a negative regulator of MAP1LC3/LC3-dependent targeting pathways (consistent with xenophagy or CASM/LAP-related ATG8ylation) that mediate macrophage immunity to Mtb. Using a targeted knockdown screen in mouse macrophages, we found that Usp15 loss increased K63-linked ubiquitination and LC3 recruitment to Mtb-associated structures, leading to reduced bacterial replication. These effects required USP15's catalytic activity and were reversed by knockdown of PRKN or inhibition of autophagy initiation. In primary human macrophages, USP15 knockdown similarly enhanced LC3 targeting and restricted Mtb growth. Importantly, pharmacological inhibition of USP15 with a selective small molecule decreased Mtb burden in human macrophages. Our findings identify USP15 as a suppressor of macrophage immunity and suggest that targeting deubiquitinases may represent a promising host-directed therapeutic strategy against tuberculosis.Abbreviations: CFU: colony-forming unit; DUBs: deubiquitinases; K48-Ub: K48-linked ubiquitin; K63-Ub: K63-linked ubiquitin; Mtb-pLux: luminescent Mtb strain Mtb; Mycobacterium tuberculosis; MOI: multiplicity of infection; NTC: non-targeting control; TB: tuberculosis.

Tuberculosis (TB) is transmitted through the air, yet the determinants of natural airborne transmission remain poorly defined. Early twentieth-century guinea pig studies demonstrated efficient airborne transmission of Mycobacterium tuberculosis (Mtb), but this paradigm has not been reestablished in contemporary containment facilities. Here, we show that ventilation can impose airflow constraints that suppress transmission under otherwise permissive conditions. Using a guinea pig model of animal-to-animal exposure, we combined transmission experiments with quantitative airflow measurements and particle transport modeling to explain why some housing configurations fail to support effective exposure. Static environments and excessive unidirectional airflow prevented transmission, whereas controlled low-velocity airflow restored evidence of exposure, including tuberculin skin test conversion, antigen-specific immune responses, and pulmonary inflammation consistent with early infection. These findings identify airflow as a critical constraint on airborne TB transmission and provide a reproducible framework to dissect host, microbial, and environmental determinants of spread.

Lung disease due to non-tuberculous mycobacteria (NTM) is rising in incidence. While both two dimensional cell culture and animal models exist for NTM infections, a major knowledge gap is the early responses of human alveolar and innate immune cells to NTM within the human alveolar microenvironment. Here we describe development of a humanized, three-dimensional, alveolus lung-on-a-chip (ALoC) model of Mycobacterium fortuitum lung infection that incorporates only primary human cells such as pulmonary vascular endothelial cells in a vascular channel, and type I and II alveolar cells and monocyte-derived macrophages in an alveolar channel along an air-liquid interface. M. fortuitum introduced into the alveolar channel primarily infected macrophages, with rare bacteria inside alveolar cells. Bulk-RNA sequencing of infected chips revealed marked upregulation of transcripts for cytokines, chemokines and secreted protease inhibitors (SERPINs). Our results demonstrate how a humanized ALoC system can identify critical early immune and epithelial responses to M. fortuitum infection. We envision potential application of the ALoC to other NTM and for studies of new antibiotics.