Development of an ultrasensitive ELISA for the detection of Mycobacterium tuberculosis antigens: an impossible challenge or a promising feat?
Date
2022
Authors
Early, Kala, author
Dobos, Karen, advisor
Mehaffy, Carolina, advisor
Schenkel, Alan, committee member
Henry, Charles, committee member
Journal Title
Journal ISSN
Volume Title
Abstract
Tuberculosis (TB) has been classically characterized as a two-state disease with active and latent phases. Latent TB infection (LTBI) is diagnosed by either the tuberculin skin test (TST) or the Interferon Gamma Release Assay (IGRA) test. However, both diagnostic tests are unable to differentially diagnose active TB and LTBI and perform poorly in immunocompromised patients. The TST is further complicated by cross-reactivity with BCG vaccination. Therefore, further diagnostic discovery for LTBI is needed for differential diagnosis and to identify those at risk of progression to active TB for subsequent treatment. Extracellular vesicles (EVs) are nanovesicles released by eukaryotic cells. EVs from TB patients contain Mycobacterium tuberculosis (Mtb) proteins, and these protein biomarkers show promise for TB and LTBI diagnostics. Our lab previously identified 31 Mtb peptides in trypsin-treated serum EVs isolated from patients with LTBI using multiple reaction monitoring-mass spectrometry (MRM-MS) methods. MRM-MS is a highly sensitive technology but is not feasible for widespread use as a diagnostic. The goal of this study was to develop an ultrasensitive ELISA against Mtb proteins for potential use as a point-of-care diagnostic. A sandwich ELISA was initially developed against Mtb proteins DnaK, Mpt32, and GroES. Reagent development for the sandwich ELISA included polyclonal antibody production using a rabbit model, murine monoclonal antibody purification and biotinylation from an existing collection of hybridoma cell lines for each antigen, and detection using a streptavidin-HRP system with a chemiluminescent substrate for signal expansion. We observed that the sandwich ELISA was complicated by non-specific binding of the DnaK and GroES antigens to the BSA block. We hypothesized that the chaperone function of these two proteins influenced them to bind to BSA. This non-specific interaction was further characterized using SPR technology and demonstrated a concentration dependent binding of DnaK to BSA. A direct-biotinylated ELISA was subsequently developed and optimized. Limit of detection (LOD) and limit of quantification (LOQ) of the direct-biotinylated ELISA was determined for each antigen: 1) GroES had an LOD of 1.959 ng/mL and an LOQ of 6.531 ng/mL, 2) Mpt32 had an LOD of 1.884 ng/mL and an LOQ of 6.278 ng/mL, and 3) DnaK had an LOD of 6.310 ng/mL and an LOQ of 21.032 ng/mL. This direct-biotinylated ELISA platform demonstrated high sensitivity with low background for all three antigens. Thus, we successfully developed and optimized an ultrasensitive ELISA for the detection of Mtb antigens.