Development of an ELISA for evaluation of swab recovery efficiencies of bovine serum albumin

Abstract

After a potential biological incident the sampling strategy and sample analysis are crucial for the outcome of the investigation and identification. In this study, we have developed a simple sandwich ELISA based on commercial components to quantify BSA (used as a surrogate for ricin) with a detection range of 1.32-80 ng/mL. We used the ELISA to evaluate different protein swabbing procedures (swabbing techniques and after-swabbing treatments) for two swab types: a cotton gauze swab and a flocked nylon swab. The optimal swabbing procedure for each swab type was used to obtain recovery efficiencies from different surface materials. The surface recoveries using the optimal swabbing procedure ranged from 0-60% and were significantly higher from nonporous surfaces compared to porous surfaces. In conclusion, this study presents a swabbing procedure evaluation and a simple BSA ELISA based on commercial components, which are easy to perform in a laboratory with basic facilities. The data indicate that different swabbing procedures were optimal for each of the tested swab types, and the particular swab preference depends on the surface material to be swabbed.

Figure 1. Sandwich ELISA setups.

Different ELISA setups (table 1, setups 1–15) were tested based on absorbance measured at 450 nm for the endpoint BSA standards (std. 80 and 0 ng/mL), and the controls without capture antibody (Ab) and primary antibody (Ab). The parameters for each setup are listed in table 1. Data is expressed as mean ± SD (n = 3).

Figure 2. Concentration curves for developed and commercial BSA ELISAs.

The optimal sandwich ELISA (table 1, setup 15) including the in-house BSA standards (std.) from 0–80 ng/mL (Developed ELISA). Capture antibody at a 1∶400 dilution, primary antibody at a 1∶800 dilution and secondary antibody at a 1∶1000 dilution. With the commercial BSA ELISA the in-house BSA standards (Commercial ELISA – BSA std.) were tested in addition to the commercial kit BSA standards (Commercial ELISA – Kit std.). There were no statistically significant difference (t-test comparing the slopes of the regression lines, p≥0.05) between the datasets of the BSA concentration curve of the developed (y = 0.0368 · x+0.686, r = 0.9853) and commercial ELISA (y = 0.0175·x+0.2144, r = 0.9534) in BSA concentration ranging from 0–40 ng/mL. Data is expressed as mean ± SD (n = 3) or mean (n = 2).

Figure 3. Recoveries from different combinations of swabbing techniques and after-swabbing treatments.

The different combinations of swabbing techniques and swab treatments were tested on a plastic surface material with gauze (in 3 mL PBS) and flocked (in 1 mL PBS) swabs. Six setups have a mean recovery >30% and they are not significantly different from the highest mean recovery (gauze cotton swab combination I/C). Data is expressed as mean ± SD (n = 3), one-way ANOVA, Dunnett’s test for post-hoc comparison vs. gauze I/C (the highest mean recovery), ns (≥0.05) not significant, ** significant at p<0.005 and *** significant at p<0.001.

Figure 4. Recovery efficiencies of BSA from seven different surfaces.

The figure shows the recovery from different surface materials expressed as a percentage of the known amount of BSA. The surfaces are swabbed with gauze cotton swabs (I/B, figure 3) and flocked nylon swabs (II/D, figure 3) respectively. Data is expressed as mean ± SD (n = 3), two-tailored, unpaired t-test, ns (≥0.05) not significant and * significant at p<0.05.