Factors Affecting Signal Generation in ELISA

To avoid the considerable time and effort required to develop and optimize a new ELISA protocol, most researchers will choose ready-to-use ELISA kits whenever possible. These kits have been optimized and validated to maximize overall signal quality for criteria such as sensitivity, specificity, precision, and lot-to-lot consistency. However, when a kit is not available for the specific target of interest or it is not optimized for the particular parameter required (e.g., sample type), laboratories may need to develop their own assay.

Specific antibodies are the essential basis of any ELISA. If matched antibody pairs exist for the target of interest, these are likely to provide the best starting point for developing or modifying an ELISA protocol to suit particular assay objectives. Otherwise, individual specific antibodies may need to be purchased and tested or new antibodies may need to be produced before an ELISA can be developed.

Nevertheless, whether one uses a manufactured ELISA kit or builds a completely new ELISA assay, it is beneficial to consider the basic factors, as outlined in Table 1, that influence the quality and intensity of signal production and its detection in ELISA. All kinds of reagents, antibodies, supplies, microplates, and instruments are available for laboratories needing to develop ELISA protocols to measure unique research targets.

Table 1. Factors that affect ELISA signal generation.

Plates used in conventional ELISA applications are typically made of polystyrene. Other materials such as polypropylene, polycarbonate, and in some instances, nylon are occasionally used. Also, some plates are gamma-irradiated to impart a positive charge, which aids coating procedures.

The absorbances of colorimetric substrates are measured by shining a laser through the base of each well. For this reason, it is essential to use a flat bottomed plate with a clear base. Fluorescent detection requires the use of an opaque black or white plate. Fluorometric plate readers measure either from above or below the plate. Chemiluminescent detection requires the same type of plate as fluorescent detection because these plates are read in the same way. Although black plates are preferred for fluorescence (background is lower) and white plates are preferred for chemiluminescence (magnifies the signal), the plates can be used interchangeably.

The most common technique for attachment of proteins to a plate is passive adsorption. This is mediated primarily by hydrophobic interactions, but some electrostatic forces may also contribute. A popular coating buffer is carbonate-bicarbonate buffer (0.2 M sodium carbonate/bicarbonate, pH 9.4). The high pH aids solubility of many proteins and peptides and ensures that most proteins are protonated with an overall negative charge, which helps when binding to a positively charged plate. Other buffered solutions such as tris-buffered saline (TBS) or phosphate-buffered saline (PBS) at physiological pH are sometimes used but coating is generally not as efficient.

In some instances, researchers require a more directed approach for attachment of the coating antibody or protein sample to the plate. Several pre-activated plates exist for this purpose. For specific, orientated binding of the coating antibody, plates that are pre-coated with Protein A or Protein G are available. These options are not advised for sandwich ELISAs because of potential cross-reactivity with detection and/or secondary antibodies. For biotinylated samples or coating antibodies, plates that are pre-coated with streptavidin are ideal.

Some assays require direct immobilization of a histidine-tagged protein, in which case nickel- or copper-coated plates are suitable. These plates can also be used to bind and orientate capture antibodies because IgG molecules have a histidine rich sequence in their Fc domain. For group-specific attachment of molecules via amines or thiols to form a covalent bond, maleic-anhydride, or maleimide-activated plates can be used. These are especially useful for direct attachment of peptide antigens, which do not coat well by passive adsorption because of their small size.

Blocking buffers consist of formulations of proteins designed to prevent non-specific binding to the plate. An optimal blocking buffer maximizes the signal-to-noise ratio and does not react with the antibodies or target protein. If cross-reactivity is observed, then a different blocking agent should be tested. If repeated cross-reactivity is observed, it may be advisable to switch to a non-mammalian protein blocking agent such as salmon serum or a protein-free blocking solution.

Some systems may benefit from the addition of a surfactant such as Tween 20 (a gentle non-ionic detergent) to the blocking solution. Surfactants can help to minimize hydrophobic interactions between the blocking protein and the antigen or antibodies. Typically a final concentration of 0.05% (v/v) Tween 20 is used. In addition, blocking buffers should be used in sufficient volumes to completely coat the wells. For example, 400 μL is generally used for each well of a 96-well plate.

The two most commonly used wash buffers in ELISA applications are Tris-buffered saline (TBS) and phosphate-buffered saline (PBS) containing 0.05% (v/v) Tween 20. To wash a plate, wells should be repeatedly filled and emptied by either aspiration or plate inversion (i.e., dumping and flicking solution into a suitable receptacle). Generally at least 3 x 5 minute washes should be applied after the incubation of coating antibody, sample, and detection antibody; 6 x 5 minute washes should be given after incubation with the enzyme conjugate. It is not necessary to wash after the blocking step, although this is not detrimental to the assay. Wash buffers should be used in sufficient volumes to completely wash the wells. For example, 400 μL is generally used for each well of a 96-well plate.

Antibodies are a key part of any ELISA workflow. For an antibody to work successfully in ELISA, it should react specifically with the antigen but not cross-react with any other component of the assay. Not all antibodies can be used successfully in ELISA applications, individual antibodies must be evaluated. For sandwich assays, where two different antibodies are required, it is essential that the two antibodies react with different epitopes on the antigen or an epitope that appears several times on the antigen.

For example, if the antigen is immobilized on the plate through the capture antibody, then the detection antibody must be able to interact with its own epitope without steric hindrance from the first antibody. In ELISA applications where a secondary antibody is used as part of the detection complex it is also essential that the capture and detection antibodies be raised in different animal species so that the secondary antibody does not react with the coating antibody. Antibodies that work well together are generally known as “matched pairs”. Most commercially-available ELISA kits use validated matched antibodies pairs, and these pairs are often available for purchase individually or as “Antibody Pair Kits”.

In addition, antibody concentrations should be considered when setting up a new ELISA. Each antibody being used will require optimization. The optimal range is partially determined by the form and origin of the antibody and also by the substrate used for signal generation. When diluting antibodies, detection antibody and enzyme conjugate, working solutions should be prepared in blocking solution to reduce non-specific interactions. For recommended coating and detection antibody concentrations see below.

Table 2. Recommended concentration ranges for coating and detection antibodies for ELISA optimization. The use of non-purified antibodies will work but may result in higher background. It is generally recommended to use affinity purified antibodies for optimal signal-to-noise ratio. Concentrations are guidelines only; for best results optimize each component individually.

The target antigen should be present in a buffer or matrix that allows it to interact with a pre-coated capture antibody or be coated to the plate directly. Direct coating may require that the antigen be exchanged into a suitable coating buffer. In rare instances the three-dimensional structure of an antigen may be altered during the adsorption process such that it no longer binds its target epitope. In such cases, the use of a plate pre-coated with a binding protein (such as a capture antibody) may eliminate this problem. If the antigen is present in the form of a biological sample the effects of the matrix (i.e. medium, serum) should be controlled by performing spike-and-recovery and linearity-of-dilution experiments.

When performing a quantitative ELISA it is essential to have an equivalent standard protein (generally a purified recombinant) whose concentration is known. The standard is used to prepare a series of solutions of known concentration by serial dilution followed by construction of a standard curve plotting concentration versus absorbance. Absorbance values for samples of unknown concentration are interpolated onto this curve to determine the actual amount of specific protein in the sample. When setting up an ELISA it is advisable to first generate and optimize a standard curve for the analyte of interest before analyzing multiple samples of unknown composition. If the standard curve displays the correct sensitivity, range and linearity, the researcher can proceed with confidence to process the samples.

The concentration of the enzyme conjugate is one of the most crucial parameters in the optimization process. The amount of enzyme that binds directly influences the amount of signal that is generated. Too little enzyme and the signal may be very weak with a poor signal-to-noise ratio. Too much and the background may be too high, again resulting in a poor signal-to-noise ratio and little distinction between standards of different concentrations. For recommended enzyme conjugate concentrations see below.

Table 3. Recommended secondary antibody concentrations for ELISA in different systems. Check the instructions for the substrate as they may recommend a more defined concentration range for the enzyme conjugate.

The choice of a particular substrate will depend on the equipment available and on the degree of sensitivity required. Chemiluminescent substrates are the most sensitive, with antigen detection possible in the sub-picogram per well range. Colorimetric and fluorescent substrates are typically able to detect low- to mid-picogram levels of antigen per well. Precipitating substrates are not used with plate assays as the precipitate settles in the wells and prevents the measurement of absorbance.

Colorimetric substrates are measured using a standard plate reader with the appropriate filters. Fluorescence is measured using a fluorometer with the appropriate excitation and emission filters. Chemiluminescence is most commonly measured using a luminometer although some plate readers have an option to read chemiluminescence or can be adapted to measure total light output. If a wavelength must be selected, measurement at 425 nm gives a crude indication of chemiluminescence. Some fluorescent plate readers can also be used if excitation is turned off.

1. John R. Crowther, Methods in Molecular Biology, the ELISA Guidebook. Second Edition. Humana Press, a part of Springer Science + Business Media, LLC 2009.
2. Butler J.E. The Behavior of Antigens and Antibodies Immobilized on a Solid Phase. In: M.H.V. Van Regenmortel, ed. Structure of Antigens. Boca Raton, FL: CRC Press, 1992: 209-259. Vol.1, 209; CRC Press, Inc.
3. Lequin, Rudolf M. "Enzyme immunoassay (EIA)/enzyme-linked immunosorbent assay (ELISA)." Clinical chemistry 51.12 (2005): 2415-2418.