Detection and Monitoring of Mycobacterium leprae Infection in Nine Banded Armadillos (Dasypus novemcinctus) Using a Quantitative Rapid Test

Abstract

Leprosy is an infectious disease caused by Mycobacterium leprae with tropism for skin and peripheral nerves. Incessant transmission in endemic areas is still impeding elimination of leprosy. Although detection of M. leprae infection remains a challenge in asymptomatic individuals, the presence of antibodies specific for phenolglycolipid-I (PGL-I) correlate with bacterial load. Therefore, serosurveillance utilizing field-friendly tests detecting anti-PGL-I antibodies, can be applied to identify those who may transmit bacteria and to study (reduction of) M. leprae transmission. However, serology based on antibody detection cannot discriminate between past and present M. leprae infection in humans, nor can it detect individuals carrying low bacillary loads. In humans, anti-PGL-I IgM levels are long-lasting and usually detected in more individuals than anti-PGL-I IgG levels. Inherent to the characteristically long incubation time of leprosy, IgM/IgG relations (antibody kinetics) in leprosy patients and infected individuals are not completely clear. To investigate the antibody response directly after infection, we have measured antibody levels by ELISA, in longitudinal samples of experimentally M. leprae infected, susceptible nine-banded armadillos (Dasypus novemcinctus). In addition, we assessed the user- and field-friendly, low-cost lateral flow assay (LFA) utilizing upconverting reporter particles (UCP), developed for quantitative detection of human anti-PGL-I IgM (UCP-LFA), to detect treatment- or vaccination-induced changes in viable bacterial load. Our results show that serum levels of anti-PGL-I IgM, and to a lesser extent IgG, significantly increase soon after experimental M. leprae infection in armadillos. In view of leprosy phenotypes in armadillos, this animal model can provide useful insight into antibody kinetics in early infection in the various spectral forms of human leprosy. The UCP-LFA for quantitative detection of anti-PGL-I IgM allows monitoring the efficacy of vaccination and rifampin-treatment in the armadillo leprosy model, thereby providing a convenient tool to evaluate the effects of drugs and vaccines and new diagnostics.

Keywords: M. leprae; PGL-I; POC; antibodies; armadillo; diagnosis; lateral flow assay; leprosy.

FIGURE 1

Flowchart of infection, treatment and sampling, Sample collected at several time points from 52 nine-banded armadillos (Dasypus novemcinctus) infected with 1 × 10⁹ M. leprae bacilli from strains: NHDP 63 (n = 33), NHDP 98 (n = 7), or Brazil 4923 (n = 11). One animal was infected in the wild. M. leprae infected armadillos that were untreated and unvaccinated (n = 17) were divided into 3 groups according to M. leprae dissemination and disease progression after inoculation: highly susceptible (within 12 months; n = 7), susceptible (within 24 months; n = 4), and resistant (after more than 36 months; n = 6); Treated animals were divided into: Armadillos vaccinated with BCG 1 month pre- (n = 6) or post- (n = 2) infection; Armadillos prophylactically vaccinated with ID93 (n = 9) or LepVax (n = 7) 1 month post-infection; Armadillos therapeutically vaccinated LepVax (n = 8) 10–32 months post-infection; Armadillos treated for 3 months with rifampin 8 or 10 months post-infection (n = 3). For the animal that was infected in the wild and LepVax vaccinated, the inoculation time (t_i) and interval between t_i were unknown. t_i, time of inoculation; t_v, time of vaccination; t_r, time of rifampin treatment; t_c, time of seroconversion (anti-PGL-I IgM ELISA > 0.45 OD_{450–background}; for the highly susceptible and susceptible group); t_m, mid-stage disease; t_l, late-stage disease; t_s, time at sacrifice (see Supplementary Table 1).

FIGURE 2

Anti-PGL-I antibody levels in M. leprae infected armadillos. Anti-PGL-I antibody levels were measured by ELISA in serum from M. leprae [1 × 10⁹ bacilli; NHDP 63 (n = 9), NHDP 98 (n = 4), or Brazil 4923 (n = 4)] infected armadillos. Sample derived from each animal is represented by the same symbol. After infection, seven armadillos disseminated before 12 months (highly susceptible; dots), four disseminated and progressed to disease within 24 months (susceptible; triangles), and six armadillos did not disseminate even 36 months post-infection (resistant; squares). ELISA results of anti-PGL-I IgM (A) and anti-PGL-I IgG (B) are displayed as optical density at 450 nm corrected for background OD values (OD_{450–background}; y-axis). The median values per group are indicated by horizontal lines. Differences between pre-infection (t_i) vs. post-infection (t_c/t_m/t_l/t_s) were determined by Wilcoxon matched-pairs test. P-values: ^∗p < 0.05. t_i, time of inoculation; t_c, time of seroconversion (for highly susceptible and susceptible animals); t_m, mid-stage disease; t_l, late-stage disease; t_s, time at sacrifice.

FIGURE 3

Correlation of anti-PGL-I IgM levels in M. leprae infected armadillos in UCP-LFA and ELISA. Serum samples (n = 41) of M. leprae [1 × 10⁹ bacilli; NHDP 63 (n = 9), NHDP 98 (n = 4), or Brazil 4923 (n = 4)] infected armadillos at multiple time points, were assessed for anti-PGL-I IgM by UCP-LFA, and for anti-PGL-I IgM by ELISA. UCP-LFA results are displayed as the Ratio value (R) between Test (T) and Flow-Control (FC) signal (x-axis); ELISA results are displayed as optical density at 450 nm corrected by background (OD_{450–background}; y-axis). R² is the square of the Spearman correlation coefficient.

FIGURE 4

Anti-PGL-I IgM levels by UCP-LFA in M. leprae infected armadillos. Anti-PGL-I IgM levels were measured by UCP-LFA in serum from M. leprae [1 × 10⁹ bacilli; NHDP 63 (n = 9), NHDP 98 (n = 4), or Brazil 4923 (n = 4)] infected armadillos divided into highly susceptible (n = 12; dots; upper panel), susceptible (n = 4; triangles; middle panel), and resistant (n = 6; squares, lower panel); sample derived from each animal is represented by the same symbol. UCP-LFA results are displayed as the Ratio value (R) between Test (T) and Flow-Control (FC) signal (y-axis). The cut-off value for discrimination between infected vs. uninfected animals is R > 0.07. The median values of each group are indicated by horizontal lines. Differences between pre-infection (t_i) vs. post-infection (t_c/t_m/t_l/t_s) were determined by Wilcoxon matched-pairs test. P-values: ^∗p < 0.05. t_i, time of inoculation; t_c, time of seroconversion (for highly susceptible and susceptible animals); t_m, mid-stage disease; t_l, late-stage disease; t_s, time at sacrifice (<12 months for highly susceptible; 12–24 months for susceptible; >36 months for resistant).

FIGURE 5

Differences in anti-PGL-I antibody levels between uninfected- and M. leprae infected highly susceptible, susceptible and resistant armadillos. Anti-PGL-I antibody levels were measured by UCP-LFA (IgM) and ELISA (IgM/IgG) in serum from armadillos infected with 1 × 10⁹ M. leprae bacilli of the following strains: NHDP 63 (n = 9), NHDP 98 (n = 4), or Brazil 4923 (n = 4). Samples were analyzed pre-infection (t_i) for all animals (n = 21) and post-infection (t_c/t_m/t_l/t_s) for highly susceptible (n = 7), susceptible (n = 4) and resistant (n = 6) armadillos UCP-LFA results are displayed as the Ratio value (R; y-axis) between Test (T) and Flow-Control (FC) signal; ELISA results are displayed as optical density at 450 nm corrected by the background for each sample (OD_{450–background}). (A) For animals for which both samples were available, significant differences between levels in pre (t_i)- vs. post (t_s)-infection sera were determined by Wilcoxon matched-paired tests; (B) Antibody levels in all sera after inoculation were determined by Mann-Whitney U-tests; P-values: ^∗p < 0.05, ^∗∗∗p < 0.001. t_i: time of inoculation; t_c: time of seroconversion (for highly susceptible and susceptible animals); t_m, mid-stage disease; t_l, late-stage disease; t_s, time at sacrifice.

FIGURE 6

Anti-PGL-I IgM levels by UCP-LFA in M. leprae infected treated/vaccinated armadillos. Anti-PGL-I IgM levels were measured by UCP-LFA in serum from M. leprae [1 × 10⁹ bacilli; NHDP 63 (n = 24), NHDP 98 (n = 3), or Brazil 4923 (n = 7)], and one animal was infected in the wild) infected armadillos (n = 35). 16 armadillos were prophylactically vaccinated with ID93 (n = 9; A) or LepVax (n = 7, B) 1 month post-infection; eight armadillos were therapeutically vaccinated with LepVax (C) 10–32 months post-infection; Eight armadillos were vaccinated with BCG (D) 1 month pre- (n = 6, black frame) or post- (n = 2, colored triangle) infection; Rifampin was provided for 3 months to armadillos (n = 3) 8 or 10 months post-infection (E). Sample derived from each animal is represented by the same symbol. UCP-LFA results are displayed as the Ratio value (R; y-axis) between Test (T) and Flow-Control (FC) signal. The median values of each group are indicated by horizontal lines. Differences in antibody levels between pre-infection (t_i) vs. post-infection (t_v/t_r/t_m/t_l/t_s) were determined by Mann-Whitney U-tests. t_i, time of inoculation; t_v, time of vaccination; t_r, time of rifampin treatment; t_m, mid-stage disease; t_l, late-stage disease; t_s, time at sacrifice.

FIGURE 7

Anti-PGL-I IgM levels by UCP-LFA in M. leprae infected armadillos after interventions. Anti-PGL-I IgM levels were measured by UCP-LFA in serum from M. leprae [1 × 10⁹ bacilli; NHDP 63 (n = 33), NHDP 98 (n = 7), or Brazil 4923 (n = 11)], and one animal was infected in the wild) infected armadillos (n = 17) divided into highly susceptible (n = 7, red dots), susceptible (n = 4, blue dots), and resistant (n = 6, green dots). 24 armadillos were prophylactically vaccinated with ID93 (n = 9, orange) or LepVax (n = 7, purple dots) 1 month post-infection; eight armadillos were therapeutically vaccinated LepVax (gray dots) 10–32 months post-infection; Eight armadillos were vaccinated with BCG 1 month pre- (n = 6, brown dots) or post- (n = 2, yellow squares) infection. Rifampin was provided for 3 months to armadillos (n = 3, teal dots) 8 or 10 months post-infection. UCP-LFA results are displayed as the Ratio value (R) between Test (T) and Flow-Control (FC) signal (y-axis). The median values of each group are indicated by horizontal lines. Differences between groups were determined by Kruskal-Wallis test with Dunn’s correction for multiple testing. P-values: ^∗p < 0.05. t_i, time of inoculation; t_c, time of seroconversion (for highly susceptible and susceptible animals); t_v, time of vaccination; t_m, mid-stage disease; t_l, late-stage disease; t_s, time at sacrifice.