Quantitative 1H Nuclear Magnetic Resonance Assay for the Rapid Detection of Pyrazinamide Resistance in Mycobacterium tuberculosis from Sputum Samples

Abstract

Tuberculosis (TB), caused by Mycobacterium tuberculosis, is one of the 10 leading killer diseases in the world. At least one-quarter of the population has been infected, and there are 1.3 million deaths annually. The emergence of multidrug-resistant (MDR) and extensively drug-resistant (XDR) strains challenges TB treatments. One of the drugs widely used in first- and second-line regimens is pyrazinamide (PZA). Statistically, 50% of MDR and 90% of XDR clinical strains are resistant to PZA, and recent studies have shown that its use in patients with PZA-resistant strains is associated with higher mortality rates. Therefore, the is an urgent need for the development of an accurate and efficient PZA susceptibility assay. PZA crosses the M. tuberculosis membrane and is hydrolyzed to its active form, pyrazinoic acid (POA), by a nicotinamidase encoded by the pncA gene. Up to 99% of clinical PZA-resistant strains have mutations in this gene, suggesting that this is the most likely mechanism of resistance. However, not all pncA mutations confer PZA resistance, only the ones that lead to limited POA production. Therefore, susceptibility to PZA may be addressed simply by its ability to form, or not, POA. Here, we present a nuclear magnetic resonance method to accurately quantify POA directly in the supernatant of sputum cultures collected from TB patients. The ability of the clinical sputum culture to hydrolyze PZA was determined, and the results were correlated with the results of other biochemical and molecular PZA drug susceptibility assays. The excellent sensitivity and specificity values attained suggest that this method could become the new gold standard for the determination of PZA susceptibility.

Keywords: Mycobacterium tuberculosis; antibiotic resistance; clinical trial; drug susceptibility test; nuclear magnetic resonance; pyrazinamide; tuberculosis.

FIG 1

¹H-NMR spectra of M. tuberculosis-positive MODS culture supernatants from the sputum samples of five different M. tuberculosis patients (samples TBCA-012, TBCA-032, TBCA-052, TBCA-068, and TBCA-95) after 3 days of incubation with PZA. The percentage of POA released was determined by integrating the PZA and POA signals.

FIG 2

Timeline for PZA drug susceptibility assays. NALC, N-acetyl-l-cysteine. (Created with Biorender.com.)

FIG 3

Kinetic curves of PZA conversion to POA monitored by ¹H-NMR-MODS in three different M. tuberculosis strains. (A) H37Rv (PZA susceptible) (in red), DM97 (PZA resistant) (in green), and a pncA knockout strain (in light blue). (B) PZA kinetics of the following four clinical sputum samples, susceptible to PZA according to the MGIT assay, the Wayne test, and pncA-Seq (WT strains): TBN-155 (in pink), TBN-177 (in yellow), TBN-211 (in black), and TBN-212 (in violet).

FIG 4

Percent PZA conversion to POA determined by ¹H-NMR-MODS associated with PZA-resistant and PZA-susceptible M. tuberculosis strains classified according to pncA sequencing, the standard Wayne test, the MGIT assay, and the CRS test. Statistical values are reported in Table S4 in the supplemental material.

FIG 5

ROC curves for the percentage of PZA hydrolyzed to POA compared against classifications performed by the MGIT assay, the standard Wayne test, pncA sequencing, and the CRS test. AUC, area under the curve.

FIG 6

Percentage of PZA hydrolyzed to POA determined by ¹H-NMR-MODS classified according to BK and MODS-GI. Only data from patients diagnosed as being positive by the CRS test were used in this analysis. The statistical data are reported in Tables S6 and S7 in the supplemental material.

FIG 7

Percentage of PZA hydrolyzed to POA determined by ¹H-NMR-MODS associated with MODS-WI groups. (A) Box plot for each MODS-WI group. The statistical values are reported in Table S9 in the supplemental material. (B) Average percentage of POA as a function of the MODS-WI. Bars represent the standard deviations.

FIG 8

Percentage of PZA hydrolyzed to POA determined by ¹H-NMR-MODS compared to the MODS-GI and pncA data for 17 patients. Sputum samples were collected on three different days of TB treatment (day 0, day 7, and day 14). In the table, the percent POA values are reported. Empty boxes correspond to samples that were not collected. All sample data are summarized in Table S10 in the supplemental material.