In scientific experiments in general and in ELISA-style binding assay in particular, proper controls are of utmost importance. Without these an interpretation of the final result is often impossible. In planning a proper plate layout, you ensure that the signals you see, are really caused by your protein interaction and not by non-specific binding (NSB is a general problem in many interaction-assay we will later spend and whole article on.)
To detect non-specific binding, you need at least these controls:
- BLANK control – Is you detection system specific for your analyte?
To detect any non-specific binding of your detection system (1st & 2nd antibody) you should incubate one coated well with just your interaction buffer (0µM analyte). You would expect to see no signal in that well and would subtract its value from the others.
- SAMPLE control – Is you analyte interacting with the block / plastic?
Often we see an unspecific interaction of analytes with the block protein or a more general trend to precipitate, thus leading to false positive signals. We will discuss later how to prevent these, but we need make sure that we detect them now. To test for these NSB events we should incubate any concentration of our analyte with wells, which have only been blocked. For example, if you test a concentration series, you have to test them once against wells with coated ligand and once against wells where nothing has been coated but blocking substance. Ideally, the wells with blocking agent only stay colourless. However, often during initial development of an assay, you will observe a concentration dependent interaction of your analyte with the block – if so, you might subtract these from your real signal. However, most times it will be better to tweak your assay to eliminate this NSB (see later article).
To really plan your plate layout we need to decide how many tests we will do. If you plan to determine KD values you need various concentrations of your analyte ranging from 1/10 to 10 times of the KD. Normally you define a dilution series to get the best results. For a typical 96 well plate you either use 7 dilutions in columns (7 values + one BLANK control [0 µM]) or 11 samples in rows. If you have a rough idea about the affinity, you can use my little ELISA simulator for getting an idea what could be the best amount of concentrations and the best dilution factor. In my research field (with typically low KD around 500nM) I typcially start with 4-10µM and make 1:4 dilutions. This leads to a dilution series spanning (4000, 1000, 250, 62.5, 15.6, 3.9, 0.9, 0.0 nM). For initial experiments single measurements might be sufficient, but for publication quality you should determine every data point at least three times.
For this assay you would coat the plate as following:
|A||Protein 1 (coat)||blocking reagent only|
And use the following interaction scheme:
|A||4000nM Protein 2 (analyte)|
Before you continue, you should be able to answer the following questions:
- Do you know how you coat your plate?
- Do you have included both controls in your plate layout (BLANK / ANALYTE control)?
In the next article, we will discuss a typcial ELISA protocol.