Table 1.
Plant pathogen detection panel.
Figure 1.
Schematic workflow depicting sequential molecular binding events of the sandwich ELISA.
Within each reaction chamber, the capture antibody is adsorbed on the reactive surface followed by surface passivation by a blocking buffer. Upon target binding to the capture antibody, alkaline phosphates (AP)-tagged detection antibody specific to the antigen is added. Addition of fluorescent substrate (PNPP or p-nitrophenyl phosphate for the traditional well format, and Attophos for the micofluidic format) activated by AP generates detectable fluorescent signal, indicating successful binding events.
Figure 2.
Coating and blocking buffer selection.
A. Different coating buffers (Optibind A-L) were tested for maximum surface binding of 11E5, 2D6, and 5E7 (n = 3). B. Blocking buffers (2% BSA, 3% skim milk, 1% casein, and Optiblock solution) were tested for Ac, WSMoV, and MYSV detection (n = 3). An ideal blocking buffer resulted in highest S/N ratios. Error bars indicate ± standard deviations.
Figure 3.
Optimization of antibody concentration.
Nine different conditions for each disease panel were tested on the microfluidic platform using combinations of three concentrations of capture Ab (11E5, 2D6, and 5E7) and three concentrations of detection Ab (MPC-AP, MYSV6-AP). Panel A–C show results for Ac, WSMoV, and MYSV detection, respectively (n = 4). The S/N ratios were plotted for each of the conditions tested. Error bars indicate ± standard deviations.
Figure 4.
Specificity of the 11E5/MPC, 2D6/MYSV6, and 5E7/MYSV6 antibody pairs was tested for Ac, WSMoV, and MYSV, respectively. The antibody pairs were tested against bacteria (Ac, SQB, Pf, and DAc) and viral (TYLCV) protein standards (n = 3). The data shows averaged S/N values with error bars representing standard deviations. The dotted horizontal line indicates the threshold (or the cutoff value) of S/N = 2 (twice of values obtained from negative controls).
Figure 5.
Study of repetitive sample loading.
Effect of repetitive sample loading on assay dynamic range was investigated, using Ac as a model. The plot indicates that multiple loading helps increase assay sensitivity (n = 3). Error bars indicate ± standard deviations.
Figure 6.
Sensitivity determination of the microfluidic platform.
Comparison of assay dynamic range for Ac (A), WSMoV (B), and MYSV (C) detection between protein standards and spiked plant extracts (n = 3) by the microfluidic platform. Error bars indicate ± standard deviations.
Table 2.
Comparison of Acidovorax avenae subsp. Citrulli (Ac), watermelon silver mottle virus (WSMoV), and melon yellow spot virus (MYSV) detection in real watermelon leaf samples by the microfluidic vs. traditional ELISA (n = 3).