Background

Mycobacterium tuberculosis (Mtb) is the causative agent of Tuberculosis (TB), the number one cause of death due to an infectious disease. TB diagnosis is performed by microscopy, culture or PCR amplification of bacterial DNA, all of which require patient sputum or the biopsy of infected tissue. Detection of mycobacterial products in serum, as biomarkers of diagnosis or disease status would provide an improvement over current methods. Due to the low-abundance of mycobacterial products in serum, we have explored exosome enrichment to improve sensitivity. Mtb resides intracellularly where its secreted proteins have been shown to be packaged into host exosomes and released into the bloodstream. Exosomes can be readily purified assuring an enrichment of mycobacterial analytes from the complex mix of host serum proteins.

Methods

Multiple reaction monitoring assays were optimized for the enhanced detection of 41 Mtb peptides in exosomes purified from the serum of individuals with TB. Exosomes isolated from the serum of healthy individuals was used to create and validate a unique data analysis algorithm and identify filters to reduce the rate of false positives, attributed to host m/z interference. The final optimized method was tested in 40 exosome samples from TB positive patients.

Results

Our enhanced methods provide limit of detection and quantification averaging in the low femtomolar range for detection of mycobacterial products in serum. At least one mycobacterial peptide was identified in 92.5% of the TB positive patients. Four peptides from the Mtb proteins, Cfp2, Mpt32, Mpt64 and BfrB, show normalized total peak areas significantly higher in individuals with active TB as compared to healthy controls; three of the peptides from these proteins have not previously been associated with serum exosomes from individuals with active TB disease. Some of the detected peptides were significantly associated with specific geographical locations, highlighting potential markers that can be linked to the Mtb strains circulating within each given region.

Conclusions

An enhanced MRM method to detect ultra-low abundance Mtb peptides in human serum exosomes is demonstrated, highlighting the potential of this methodology for TB diagnostic biomarker development.

Identification of biomarkers for latent Mycobacterium tuberculosis infection and risk of progression to tuberculosis (TB) disease are needed to better identify individuals to target for preventive therapy, predict disease risk, and potentially predict preventive therapy efficacy. Our group developed multiple reaction monitoring mass spectrometry (MRM-MS) assays that detected M. tuberculosis peptides in serum extracellular vesicles from TB patients. We subsequently optimized this MRM-MS assay to selectively identify 40 M. tuberculosis peptides from 19 proteins that most commonly copurify with serum vesicles of patients with TB. Here, we used this technology to evaluate if M. tuberculosis peptides can also be detected in individuals with latent TB infection (LTBI). Serum extracellular vesicles from 74 individuals presumed to have latent M. tuberculosis infection (LTBI) based on close contact with a household member with TB or a recent tuberculin skin test (TST) conversion were included in this study. Twenty-nine samples from individuals with no evidence of TB infection by TST and no known exposure to TB were used as controls to establish a threshold to account for nonspecific/background signal. We identified at least one of the 40 M. tuberculosis peptides in 70 (95%) individuals with LTBI. A single peptide from the glutamine synthetase (GlnA1) enzyme was identified in 61/74 (82%) individuals with LTBI, suggesting peptides from M. tuberculosis proteins involved in nitrogen metabolism might be candidates for pathogen-specific biomarkers for detection of LTBI. The detection of M. tuberculosis peptides in serum extracellular vesicles from persons with LTBI represents a potential advance in the diagnosis of LTBI.