Double-Sandwich ELISA Protocol

Introduction of Double-Sandwich ELISA Protocol

The enzyme-linked immunosorbent assay (ELISA) has enjoyed application in many areas because of its high specificity and sensitivity. Many variations are known including the indirect ELISA, competitive ELISA, the antibody-sandwich ELISA, antibody-capture ELISA, and the double antibody-sandwich ELISA. One of the most useful immunoassay formats is the sandwich ELISA designed for detection of soluble antigens. The difference between a capture ELISA and a sandwich ELISA is that a capture ELISA is designed to measure the amount of antibody present in a sample (typically a serological test), while a sandwich ELISA measures the amount of antigen in the sample. Thus, for a sandwich ELISA, a pair of antibodies to the target antigen is needed. The antigen is trapped between the capture antibody and the detector antibody. The example shown represents a direct binding sandwich ELISA since the detector antibody is directly labeled with an enzyme. A common variant is the indirect sandwich ELISA where the binding of the detector antibody is visualized by binding of a third antibody that is conjugated to the reporter molecule (i.e., if the detector antibody is made in a goat, the third or “reporter antibody” would be an enzyme-conjugated anti-goat antibody). In this scenario, the primary capture antibody could not be produced in goats but could be a mouse monoclonal antibody.

The detector and/or reporter antibody can be labeled with an enzyme and the sandwich detected with suitable colorimetric, fluorescent, or luminescent substrates. Alternatively, the detector antibody can be labeled with biotin and the sandwich detected using enzyme-conjugated streptavidin. The basic requirement of a sandwich ELISA (unless a repetitive epitope exists on the target antigen) is that two antibodies binding different epitopes on the same antigen are required. Furthermore, binding of one antibody must not interfere with binding of the second antibody even if they bind different epitopes.

Figure 1. Schematic of typical sandwich ELISAs.

Since the performance characteristics (specificity and sensitivity) of a sandwich ELISA are directly related to the quality of the antibodies (their binding affinities, stability, etc.). Finding matched antibody pairs, especially two monoclonal antibodies (mAbs), with these properties is often difficult. Monoclonal antibodies offer many advantages versus polyclonal antibodies including identification of the antibody-binding epitope, improved assay specificity and reliability since a positive result requires binding of two highly specific reagents with known epitopes, the ability to select high- affinity conformational antibodies with the desired binding specificity, and a consistent source of high-quality reagent. The sandwich ELISA itself has many advantages including that the sample need not be purified before analysis. Antigen purification and concentration is accomplished during the capture phase of the assay resulting in a three- to five fold (or greater) increase in sensitivity versus a direct or indirect ELISA. Sandwich ELISAs usually display high specificity and greater confidence in the result because target detection requires binding of two distinct antibodies. However, their ability to detect low levels of target in complex samples makes them ideal tests for measuring the presence of the target antigen in unknown samples, e.g., food, environmental, or clinical samples.

Research in our laboratory has focused on development of sensitive immunoassays for botulinum neurotoxin (BoNT). Botulinum neurotoxins are produced by the anaerobic bacterium Clostridium botulinum. These toxins, as with many bacterial toxins, are large, complex di-chain protein toxins. In the case of BoNT, the active molecule consists of a 100 kDa heavy chain and a 50 kDa light chain joined by a single disulfide bond. Our initial approach was to screen hybridoma cell fusion supernatants using traditional indirect ELISA. Neurotoxin was adsorbed onto the bottoms of 96-well microassay plates. Supernatants from cell fusion experiments were added and antibody binding detected using an anti-mouse antibody conjugated to horseradish peroxidase (HRP). Earlier studies to develop monoclonal antibody-based sandwich ELISA for detecting BoNT serotype B were only partially successful since the majority of the mAbs isolated failed to bind the toxin in solution. All of the mAbs isolated displayed strong binding in the indirect ELISA (impressive titration curves), good specificity (i.e., binding serotype B but not serotype A), and strong binding on Western blots. However, none of these antibodies bound the toxin in solution; hence, none were suitable as capture antibodies for a sandwich ELISA. A discouraging result after expending significant resources to generate the hybridoma clones, purify the mAbs and label with biotin in order to identify suitable antibody pairs for a sandwich ELISA. Adsorbing protein antigens onto the plastic surface of microassay wells can result in modifications of the protein. Many changes can be envisioned including changes to the surface charge of the protein, mild denaturation resulting in exposure of cryptic epitopes, and blocking surface epitopes. Studies in our laboratory suggested that BoNT was modified when absorbed onto the plastic bottoms of microassay wells. Similar results were observed with two different BoNT serotypes and with nontoxic BoNT-associated proteins. Interestingly, the BoNT antibodies could be induced to bind toxin in solution by treating the toxin with weak SDS solutions or adjusting the pH. However, these steps necessitate additional sample preparation and only marginally improved assay performance. The double-capture ELISA outlined was applied as a screening tool to evaluate large numbers of hybridomas supernatants following cell fusion experiments in order to select for mAbs able to capture toxin in solution. Using this screen, it has been possible to select useful antibody pairs and construct sandwich ELISAs for both BoNT serotypes B, E, and F (latter two unpublished) as well as good antibody pairs for the nontoxic BoNT hemagglutinin-70-associated protein.

Once mAbs suitable for a sandwich immunoassay are identified, they can be formatted into numerous different tests, e.g., sandwich ELISA, lateral-flow devices, and various immunobiosensors. The critical reagents in these different assay formats are the antibodies.

Methods

All procedures are carried out at room temperature unless otherwise indicated. Preparation of BoNT dilutions, additions to assay plates, and plate washing performed in a biosafety cabinet.

Preparation of 96-Well Microassay Plates

1. Prepare primary capture antibody-coated microassay plates as follows. Add 100 μL/well of a 1 μg/mL solution of goat anti- mouse IgG Fc γ fragment prepared in 0.05 M carbonate buffer (pH 9.6) to black, flatbottom, Maxisorp 96-well microassay plates. Cover the plates with 96-well plate lids or plastic film sheets. Incubate the plates at 4 °C overnight or until needed (see Notes 4 and 2).
2. Aspirate the primary capture antibody solution and block remaining reactive sites by adding 300 μL per well of a solution of 5 % nonfat dry milk prepared in TBST buffer. Cover the plate and incubate at 37 °C for 1 h (see Note 5).
3. Aspirate the block solution and add 100 μL (as little as 50 μL can be used) per well of hybridoma supernatant obtained from 96-well cell culture plates following a cell fusion experiment. The plates are then covered and incubated at 37 °C for 1 h (see Note 6).
4. The hybridoma supernatant is aspirated and the plates washed three times with TBST using an automatic plate washer (see Note 7).
5. Add antigen to the plate. For BoNT, we add 50 μL of a 200 ng/mL solution in blocking buffer. The plates are then sealed with adhesive plastic fi lm, placed in a secondary containment tray, and incubated 1 h at 37 °C (see Note 8 ).
6. The plates are next washed a minimum of three times with TBST buffer using an ELx405 plate washer in a BSC.
7. Next 50 μL of the secondary detection antibody, in this example a rabbit polyclonal anti-BoNT diluted 1:5,000 (when the stock solution is at 1 mg/mL), is added to the plates. The plates are sealed with plastic films and incubated at 37 °C.
8. The plates are then washed as in step 6 above.
9. Next 50 μL of a 1–5,000 dilution of an HRP-conjugated goat anti-rabbit IgG (whole molecule), affinity (isolated, antigen - specific antibody is added to each well. The plates are then covered and incubated at 37 °C for 1 h.
10. The plates are then washed a total of 6 times with TBST as in step 6 above.
11. For evaluation of primary cell fusion plates, 50 μL SuperSignal West Pico chemiluminescent substrate prepared as suggested by the manufacturer is added to the plates. The plates are then incubated at room temperature with shaking for 3 min. The luminescent signal is recorded with a 96-well plate luminescence reader set for 0.1 s per well. Data is captured and analyzed with Excel (see Note 9). Results from a typical double-capture ELISA screen of 96-well hybridoma cell culture plates.