Ubiquitin-mediated xenophagy, a type of selective autophagy, plays crucial roles in host defense against intracellular pathogens including Mycobacterium tuberculosis (Mtb). However, the exact mechanism by which host ubiquitin targets invaded microbes to trigger xenophagy remains obscure. Here we show that ubiquitin could recognize Mtb surface protein Rv1468c, a previously unidentified ubiquitin-binding protein containing a eukaryotic-like ubiquitin-associated (UBA) domain. The UBA-mediated direct binding of ubiquitin to, but not E3 ubiquitin ligases-mediated ubiquitination of, Rv1468c recruits autophagy receptor p62 to deliver mycobacteria into LC3-associated autophagosomes. Disruption of Rv1468c-ubiquitin interaction attenuates xenophagic clearance of Mtb in macrophages, and increases bacterial loads in mice with elevated inflammatory responses. Together, our findings reveal a unique mechanism of host xenophagy triggered by direct binding of ubiquitin to the pathogen surface protein, and indicate a diplomatic strategy adopted by Mtb to benefit its persistent intracellular infection through controlling intracellular bacterial loads and restricting host inflammatory responses.

GeneXpert MTB/RIF (Xpert) assay has been endorsed for the diagnosis of pulmonary TB due to its high sensitivity and specificity for culture positive TB. There is no doubt that Xpert could not be more sensitive than mycobacterial culture, while the positive rate of Xpert among sputum samples was higher than that of mycobacterial culture in our laboratory. We therefore carried out a prospective study to determine a potential explanation for this unexpected result regarding the clinical use of Xpert. Overall, a total of 558 patients meeting inclusion criteria were enrolled in final analysis between August 2017 and September 2017 in Beijing Chest Hospital. The overall positive rate of Xpert among sputum samples was 45.9% (256/558), which was significantly higher than that of liquid culture (33.4%, 184/558; P < 0.01). The percentage of culture negative result in salivary sputum was significantly higher than that in mucoid sputum [odds ratio (OR): 5.04, 95% confidence interval (95% CI): 2.74–9.28; P < 0.01]. In addition, the TB cases having previous treatment history had a higher proportion of culture negative result than new cases (OR: 4.26, 95% CI: 1.61–11.28; P = 0.01). In conclusion, the results of this study demonstrate that Xpert outperforms mycobacterial culture in detecting MTB from salivary sputum. In addition, the previously treated patients are more likely to yield negative culture results. Our data will provide important hints to formulate an appropriate diagnostic algorithm for pulmonary tuberculosis based on the appearance of sputum samples.