Combining the specificity of antibodies with a detection system of some kind – in the right way – can yield one of the most powerful and reliable techniques for the detection or measurement of an analyte (the ligand) in biological samples: the Enzyme-Linked Immuno Sorbent Assay, or ELISA.
At its most basic an ELISA measures a signal, typically developed by an enzyme conjugated to a detection antibody cleaving a substrate molecule. The more ligand present, the more antibody binds to it and the more signal is developed.
While the principle is straight forward, they can be devilishly hard and time consuming to develop; selecting the right reagents, optimising coating/detection antibody concentrations, incubation times, washing steps, establishing the dynamic range of the assay and the effects of different sample matrices.
Whether being used to qualitativly, quantitativly or semi-quantitativly measure an analyte – the ligand – all ELISAs use antibodies to either capture or detect (or both) an analyte of interest. But there are different ways of going about this.
In general for all ELISAs a signal is developed using a reporter molecule or enzyme attached to a detection antibody. The majority of ELISAs tend to use a horseradish peroxidase (HRP) conjugated detection antibody allowing the use of a number of different substrates to develop the signal, such as TMB or ABTS, but there are many other options available.
Samples containing the analyte (antigen) of interest are added to the plate and allowed to adsorb to the plate surface. The plate is then washed to remove unbound sample, and a detection antibody, specific for the antigen of interest is added to the plate. The detection antibody is directly conjugated to a reporter molecule or an enzyme such as HRP allowing a signal to be developed.
Similar to a direct ELISA, except that the detection antibody is not directly conjugated to a reporter molecule. Instead a labelled secondary antibody that binds the detection antibody is used to indirectly detect the presense of the antigen or protein of interest.
Samples are inclubated in the presence of a fixed concentration of a labelled version of the analyte of interest. The analyte in the sample then competes for binding to a fixed quantity of a detection antibody. As the level of analyte ina sample increase, this reduces the available binding sites for the fixed concentration of labelled analyte, therefore a lower assay signal indicates a higher amount of analyte in the sample.
Sandwich ELISAs are generally considered one of the most robust forms of ELISA, requiring little or no sample preparation before use. A typical sandwich ELISA uses a 'capture' antibody pre-coated on to a microplate, samples are added and the analyte captured. The plate is then washed to remove unbound sample components, and a labelled detection antibody (often, but not necessarily, polyclonal) added, 'sandwiching' the analyte between these two antibodies.
Most labs will have everything needed to run an ELISA, and most plate readers will come with some kind of analysis software, however, if you don't have appropriate analysis software and are looking for some we would recommend SoftMax Pro. The package has optional security and validation modules for use in regulated labs.
If you don’t need to run ELISAs frequently enough to buy a new software package, there are a few free web-based options available too, for example here.
Fitting the right curve model to your data can be one of the most important steps in analysing and obtaining relaible, reproducible, results with a quantitative ELISA. Generally speaking, an ELISA calibration curve would give data points descriptive of a sigmoidal (S-shaped) or logarithmic curve. Either type of curve is usually best described using a four or five parameter logistic curve fit, though an appropriate curve fit for each kit will usually be recommended by the manufacturer.
It's easy to get caught out by assuming the methodology for this ELISA will be like the other one. This leads to one of two things, running around the lab manically half way through an assay when you've realised, or terrible data and a failed assay at the end of the day.
Don't rely on your memory for this! Many an assay has had to be repeated simply because people couldn't be sure which sample was which later on. Many kit booklets include a plate map, so you can use this and keep the kit booklet stapled in your lab book or alternatively download one from HERE. It's also a good place to make a note of the kit lot number as well. Don't forget to include the sample dilution, if any.
Generally speaking, to have confidence in your results, duplicates are best. Discount and reanalyse samples with high %CV (many labs set a threshold of about 20-30%).
Reverse pipetting is fine for adding reagents like detection antibodies or substrates to the plate. If possible, use a multichannel when adding a single reagent to the entire plate.
To avoid touching the bottom of the assay plate and disturbing/scratching any coated reagents, which might end up giving you odd results, and for consistency.
The reaction products of TMB can precipitate out of solution, resulting in a visible dark black/brown spots on the bottom of a plate, and, sometimes, crazy results.
We are dedicated to providing the best reagents and service in the UK. If you have any questions or comments CLICK HERE to leave your contact details and enquiry. We will get back to you within 24 hours.