f Point-of-care (POC) diagnostic testing of tuberculosis (TB) is a tremendous unmet need. In this study, four urinary mycobacterial antigens were identified through two independent approaches using IgG capture and immunodepletion methods. Among these, ModC was validated by a multiple reaction monitoring (MRM) method. As expected, the biomarkers elevated the clinical validity of TB diagnosis when combined with preexisting markers.
The high morbidity and mortality rates of mycobacterial tuberculosis (TB) necessitate the development of a reliable and convenient diagnostic system with easy accessibility and high diagnostic performance (1). Because urine is a top-priority biomarker source for point-of-care (POC) testing, systematic approaches for identifying tuberculosis antigens in human urine samples have recently been made (2-4). Lipoarabinomannan (LAM) is one of the best known biomarkers found in urine, and the assay for detecting LAM in the urine of TB patients is commercially available (5-7). The LAM tests show a wide range of sensitivity and specificity and have particular clinical utility for the diagnosis of TB in HIV-infected patients (7-9). Recently, the Campos-Neto group reported the identification of TB antigens in urine samples, but the biomarker reportedly fell short of its expected performance. with a diagnostic sensitivity of less than 50% (4).There is still an urgent need to identify more sensitive and specific biomarkers in urine for TB diagnosis. We aimed to identify potential biomarkers in urine through a unique approach whereby IgG was used to capture mycobacterial antigens in urine samples. We thought that patients' IgGs displaying differential binding patterns toward TB antigens could be used to capture the counterpart antigens excreted into the urine. For this, serum samples were collected from patients with active pulmonary tuberculosis, and IgG molecules were purified on a protein G column. The IgG fractions were clearly resolved from those of the rest of the serum proteome, where the purity of IgG molecules in fraction B was Ն 95% as assessed by an electropherogram of stained SDS-PAGE gels (data not shown).To test whether the IgG fractions contained antimycobacterial antibodies, binding patterns were analyzed by Western blotting using the purified IgG as a primary antibody. First, Mycobacterium tuberculosis H37Rv cells were cultured, and all lysates and secretomes were obtained. After heat denaturation, the lysates were separated into buffer-soluble and insoluble fractions and immunoblotted against IgG samples purified from the healthy individuals (Fig. 1A) or the TB patients (Fig. 1B). Representative stained SDS-PAGE gels are shown in Fig. S1A in the supplemental material. The overall pattern was that the IgG samples from the TB patients were more responsive to binding with mycobacterial proteins than those from the healthy individuals. The healthy individuals seemed to possess responsive antibodies, which may be explained by inferring that humans have innately developed a repertoire of a...