First of all let me say the technique of labeling tissues (immunohistochemistry, IHC), and cells (immunocytochemistry, ICC) is indeed immunoscience NOT alchemy, though at times it may certainly seem like alchemy!

But to scientists inexperienced in this technique, who typically see the results of IHC/ICC experiments in the form of pretty pictures, it can certainly seem like alchemy.

While these pictures can be pretty in their own right, the pictures display a range of important scientific information. The illusion of seeing such a pretty picture is the observer often thinks the process, time and skill involved in obtaining these images is relatively little and easy to do. Nothing could be further from the truth. A good portion of IHC/ICC experimental conditions is all determined empirically. Trial and “error” is part of all scientific research, and IHC/ICC is no exception.

The purpose of this article is to give the beginner a list of some initial variables one must think about before attempting his/her first set of IHC/ICC experiments. I have one caveat: there is perhaps an exception to most of what I may say below. Remember, in research as in life, not everything is an absolute.

So here’s what you need to achieve gold (rather than lead) standard results from IHC/ICC…


In real estate investing, a famous phrase which describes the potential investing success, is often “location, location, location”. The same can be said for IHC/ICC, “fixation, fixation, fixation“.  This step is the first and perhaps the most important major step in obtaining successful results in IHC/ICC experiments. The wrong fixation protocol will almost guarantee you get undesirable results, that is, no signal from the protein you are trying to observe in your sample. Fixation protocols themselves have a wide range of variables, such as:

  • Chemicals used to fix the sample (paraformaldehyde, methanol, acetone mixtures and others!)
  • Temperature (4° C to 37° C) and
  • Time (15 min to overnight) for which a sample should be fixed

Primary Antibody Selection

Will you use monoclonal or polyclonal antibodies, when should you use one over another? Who should you buy from? I’ll cover this in a future article.  Beware: the answers can be lengthy!


There are two types of controls one should use, a negative control and a positive control. Each control tells you something different, and is used to help insure an accurate interpretation of your results.

A negative control, in which you are using a primary antibody and a secondary antibody in the experimental group, consists of the omission of the primary antibody in the negative control. This negative control will give you an idea of how much signal, if any, is due to the secondary antibody alone.

Another type of negative control would substitute the primary antibody for non-immune serum or even better, pre-immune serum, followed by detection with the secondary antibody.

In the case of ICC, where the experimental group is cells transfected to express your protein for example, using parental cells should yield no specific signal due to the primary antibody.

A positive control tells you two things: whether your technique is sound, and whether the primary is detecting the appropriate target. Positive controls can come in many forms depending upon the type of experiment. In the case of IHC, a separate slide with tissue known to be immunopositive for your protein of interest should be used.

You have to remember, if you are labeling your protein of interest with an antibody, you may observe background labeling. The background may be the result of binding by the primary, secondary or both. Antibodies are inherently “sticky” proteins by nature, so an antibody may not necessarily bind to only the proper epitope, thus generating a false positive. If you have the proper controls in place, these will only make your results more compelling should you get a true signal.

Signal Detection

In short, will you use a chromagenic substrate, which requires brightfield microscopy to observe the staining pattern, or will you use fluorophores commonly conjugated to antibodies to see the fluorescent signal? The answer depends on your needs.

Do you need to see your sample’s structure and only one protein? If so, then use brightfield. Or do you need to label two or more proteins for colocalization purposes? If so, then use a microscope equipped for fluorescence.

In the age of computers it is possible with the right equipment to overlay a fluorescent image onto a brightfield image, thus maintaining the ability see both cellular structure and protein localization. However, if you are only using a single label and require the sample’s structure, you will typically achieve better results by using brightfield microscopy.


With the recent advances in microscopy it is possible to use a variety of microscopy technologies to observe your results. If you have a core instrument facility, and you aren’t sure what to use even after reading and asking friends, I recommend speaking to the director of the facility for advice.

Generally speaking, people get sharper images with laser scanning confocal microscopes than they do with compound microscopes. However, to be fair, this is dependent on your sample as well. I have heard from researchers who have used a compound microscope equipped for fluorescence getting better images from their samples than they do on a confocal microscope, so know your sample!

Image Processing

Many people use a variety of software for post-processing of images which are typically digitally acquired. One essential tip is to always maintain the original image file. Never make changes to the original image by saving over it, save the changes as a new file. Once you lose the original you are in trouble should you ever need to reproduce the original, especially to a scientific editorial board, or for a university/federal investigation.

In addition, before you make changes to the images, decide which journal(s) you are going to submit to and see what their criteria are for the acceptable altering of original images.

Some journal publishers now have forensic image analysis teams and all submitted images are put through special software which can identify what types of changes were made to the image. Scientists have been asked to reproduce the originals, or simply go back and alter the original image with only the accepted types of changes allowed by the journal.

IHC/ICC is a really useful technique but, like all useful techniques, only if it is carried out with due care and attention. I’ve outlined here some of the variables you need to consider before starting your experiments to ensure that your efforts bring in a pile of golden results.

Do you have any IHC/ICC tips or questions?

Originally published November 2, 2009.  Revised and updated April 4, 2016.