Early detection of Mycobacterium avium subsp. paratuberculosis infection in cattle with multiplex-bead based immunoassays

Abstract

Johne's Disease (JD), caused by Mycobacterium avium subspecies paratuberculosis (MAP), results in significant economic loss to livestock production. The early detection of MAP infection in animals with extant serological assays has remained challenging due to the low sensitivity of commercially available ELISA tests, a fact that has hampered the development of effective JD control programs. Our recent protein microarray-based studies identified several promising candidate antigens that are immunogenic during different stages of MAP infection. To evaluate these antigens for use in diagnostic assays and reliably identify animals with MAP infection, a multiplex (Luminex®) assay was developed using color-coded flourescent beads coupled to 6 MAP recombinant proteins and applied to screen 180 serum and 90 milk samples from cows at different stages of MAP infection including negative (NL), fecal test positive/ELISA negative (F+E-), and fecal positive/ELISA positive (F+E+). The results show that while serum antibody reactivities to each of the 6 antigens were highest in F+E+ group, antibody reactivity to three of the six antigens were identified in the F+E- group, suggesting that these three antigens are expressed and provoke antibody responses during the early infection stages with MAP. Further, antibodies against all six antigens were elevated in milk samples from both the F+E- and F+E+ groups in comparison to the NL group (p<0.01). Taken together, the results of our investigation suggest that multiplex bead-based assays are able to reliably identify MAP infection, even during early stages when antibody responses in animals are undetectable with widely used commercial ELISA tests.

Fig 1. Distribution of serum multiplex assay median fluorescent intensity (MFI) to each antigen among groups.

The violin plots (gray-filled) show the distribution shape of the data among NL (n = 60), F+E- (n = 60), and F+E+ (n = 60). The box plots in the center represent the interquartile range. The vertical line on each box represents 1.5x interquartile range (IQR), and the dots represent outliers. The symbol * indicates p < 0.05 when MFI in infected groups (F+E- or F+E+) compared to MFI in NL group, and ** indicates p < 0.01 based on Mann-Whitney’s U test.

Fig 2. Distribution of milk multiplex assay MFI to each antigen among groups.

The violin plots (gray-filled) show the distribution shape of the data among the NL (n = 30), F+E- (n = 30), and F+E+ (n = 30) groups. The box plots in the center represent the interquartile range. The vertical line on each box represents 1.5x interquartile range (IQR), and the dots represent outliers. The symbol ** indicates p < 0.01 based on Mann-Whitney’s U test when MFI in infected groups (F+E- or F+E+) compared to MFI in NL group.

Fig 3. Comparison of serum antibody reactivity of multiplex and ELISA to recombinant proteins.

ROC curves of serum multiplex reactivity to 6 recombinant MAP proteins were compared with those of serum ELISA using the same recombinant antigens (NL n = 30, F+E+ n = 60). The red ROC curves represent data from serum ELISA and blue ROC curves represent data from multiplex assay. The tables inside the plots describe the name of antigen, sensitivity, specificity and AUC.

Fig 4. Comparison of milk multiplex and ELISA antibody reactivity.

Milk multiplex antibody reactivity to recombinant MAP proteins was compared with IDEXX ELISA test results in the F+E- (n = 30) and NL (n = 30). (A) Table of AUC, cutoff, Sensitivity and Specificity; (B) ROC curves.