Screening for Mycobacterium tuberculosis Infection Using Beijing/K Strain-Specific Peptides in a School Outbreak Cohort

Abstract

Background: The Beijing strain of Mycobacterium tuberculosis (M. tb) has been most frequently isolated from TB patients in South Korea, and the hyper-virulent Beijing/K genotype is associated with TB outbreaks. To examine the diagnostic potential of Beijing/K-specific peptides, we performed IFN-γ release assays (IGRA) using a MTBK antigen tube containing Beijing/K MTBK_24800, ESAT-6, and CFP-10 peptides in a cohort studied during a school TB outbreak.

Methods: A total of 758 contacts were investigated for M. tb infection, and 43 contacts with latent TB infection (LTBI) and 25 active TB patients were enrolled based on serial screening with QuantiFERON-TB Gold In-Tube tests followed by clinical examinations. Blood collected in MTBK antigen tubes was utilized for IGRA and multiplex cytokine bead arrays. Immune responses were retested in 24 patients after TB treatment, and disease progression was investigated in subjects with LTBI.

Results: Total proportions of active disease and LTBI during the outbreak were 3.7% (28/758) and 9.2% (70/758), respectively. All clinical isolates had a Beijing/K M. tb genotype. IFN-γ responses to the MTBK antigen identified M. tb infection and distinguished between active disease and LTBI. After anti-TB treatment, IFN-γ responses to the MTBK antigen were significantly reduced, and strong TNF-α responses at diagnosis were dramatically decreased.

Conclusions: MTBK antigen-specific IFN-γ has diagnostic potential for differentiating M. tb infection from healthy controls, and between active TB and LTBI as well. In addition, TNF-α is a promising marker for monitoring therapeutic responses. These data provide informative readouts for TB diagnostics and vaccine studies in regions where the Beijing/K strain is endemic.

Keywords: Beijing/K strain; IFN-γ release assay; Mycobacterium tuberculosis; cytokine; latent tuberculosis infection; outbreak.

Figure 1

Flowchart of M. tb infection screening during the TB outbreak and recruitment of study subjects. A total of 28 active patients with TB and 70 individuals with LTBI were found based on contact screening tests for M. tb infection in 758 individuals. Among the contacts, 25 TB patients and 43 subjects with LTBI were recruited for this study. Immune responses were followed up in 24 patients with TB and 30 subjects with LTBI after successful treatment. * MTBK antigen tubes were used.

Figure 2

M. tb genotypes in clinical isolates. M. tb genotypes were confirmed by PCR in 6 of 10 patients who showed M. tb growth in culture. All clinical isolates were Beijing/K strains of M. tb. M: 100 bp DNA ladder, Lane 1: Beijing strain (~ 350 bp), Lane 2: K strain (~ 250 bp), Lane 3: M strain (~ 150 bp), Lane 4: Non-Beijing strain (~ 500 bp), Lane 5: negative control.

Figure 3

IFN-γ responses to the Beijing/K-peptide combination and diagnostic accuracy of identifying M. tb infection. IFN-γ responses to MTBK antigen were significantly higher in active TB and LTBI compared with HC. M. tb infection and disease status were differentiated by IFN-γ responses. The median value of IFN-γ is marked in red, and a dotted line represents the cut-off for positive results (0.065 IU/mL) (A). Immune responses to the MTBK antigen had good diagnostic values (AUC > 0.7) for distinguishing between active TB and LTBI (D), as well as between infected and non-infected individuals (B, C). *P < 0.05, **P < 0 .01, ***P < 0.001 by one-way ANOVA with subsequent Kruskal-Wallis test, HC, healthy control.

Figure 4

Measurement of multiple cytokine responses to Beijing/K-combination. Among IFN-α, IL-13, IL17, CXCL10, CCL5, and TNF-α, the IL-13 and CXCL10 responses were significantly higher in LTBI and active TB compared with HC (antigen minus nil). However, disease status (active TB vs. LTBI) was not differentiated by the cytokine/chemokine responses. The median responses are marked by red lines. *P < 0.05, ***P < 0.001 by one-way ANOVA with subsequent Kruskal-Wallis test, HC, healthy control.

Figure 5

Follow-up immune responses to the Beijing/K-peptide combination after TB treatment. (A) Distribution of cytokines and chemokines as immune responses to the Beijing/K-peptide combination after TB treatment. The median responses are marked by red lines. (B) Longitudinal analysis of cytokine and chemokines after TB treatment. Significant decreases in IFN-g response to the MTBK antigen were observed in 24 TB patients who finished anti-TB treatment. Other cytokines and chemokines tested in this study did not show significant changes after treatment (antigen minus nil). *P < 0.05 by Wilcoxon signed rank test, Tx, treatment.

Figure 6

Treatment efficacy measured by TNF-α. TNF-α concentration (nil) was significantly reduced after TB treatment (A, B). The proportion of responders with a high level of TNF-α decreased, while that of responders with a low level of TNF-α dramatically increased (C). The horizontal red line represents the median value of TNF-α. ***P < 0.001 by Wilcoxon signed rank test, Tx, treatment.