The visualisation of proteins or organelles in cells and other complex biological systems is ‘bread and butter’ stuff for cellular and molecular biologists; it’s performed day-in, day-out in labs across the globe. But this doesn’t mean that the most popular approaches currently used for protein visualisation – dye staining, antibody labelling and fusion protein over-expression systems – are ideal. Almost every scientific technique has its advantages and drawbacks, and the various options currently available for protein visualisation are not exceptions to this rule, so let’s have a look at the pros and cons of each.
The traditional approach is to use dyes that target specific organelles or cellular structures – histologists have been doing it for centuries. Although the scope of this technique is generally limited to cellular macrostructures due to the relatively low specificity of most dyes, it does allow effective visualisation using light microscopy. However, this approach often requires significant sample preparation, and cannot be performed on live cells. There can also be issues with dye specificity and penetration if the environmental conditions (pH, ion concentrations, etc.) change significantly from the expected ‘normal’ parameter during sample preparation.
Immunolabelling uses target-specific antibodies, usually in combination with fluorescent tags, to detect and localise proteins of interest. Although there are many variations of this technology, they can be divided into two general techniques – direct and indirect. For direct immunolabelling, the tag is bound directly to the target-specific antibody. However, because of the high cost of fusing a tag to antibodies targeting each protein of interest, an indirect method is more commonly used. For this technique, a recognition antibody from a non-native source – for example, a mouse-origin antibody when investigating human cell lines – binds the target protein, then a tagged generic antibody (anti-mouse in the example) targets and binds the recognition element.
HAP1 and HAP1 cells gene-edited to knockout SLC30A6 (HAP1_SLC30A6, catalogue number: HZGHC002784c010) with the HPA antibody HPA057328 targeting SLC30A6 demonstarting the specificity of this antibody. The samples were prepared in parallel using the same antibody dilutions and reagents, and both images are acquired with the exact same settings. Images curtesy of Dr Emma Lundberg, Cell Profiling facility. KTH Royal Institute of Technology.
Regardless of which approach is used, the most obvious disadvantage of immunolabelling is that it cannot be used for live cell imaging. On the contrary, samples often require extensive preparation protocols with multiple, laborious wash steps before visualisation. Another major drawback of immunolabelling techniques is the specificity of the antibodies used. Although a large number of companies now offer antibodies to target any protein of interest, characterisation of these products is often poor, with little information available on specificity and potential off-target binding activities. Recent reports suggest that up to 50% of commercially available antibodies are do not perform satisfactorily, either due to problems with cross-reactivity or lack of sensitivity1.
An alternative tactic is to attach a label directly to the protein of interest. By creating fusion proteins composed of the target protein and a fluorescent or luminescent reporter, researchers can more easily follow the intracellular movement of these proteins in live cells. Over-expression systems are often employed, using a plasmid vector to introduce the fusion protein gene, and a promoter which is controls the expression of the introduced protein. These models are well understood, and are widely used for protein production in the biotechnology and biopharmaceuticals sectors. However, they can only be used to follow the engineered, non-native version of the target protein. The behaviour if the introduced protein may vary from the original target as expression is controlled independently to the regulation mechanisms of the native protein. Furthermore, if the introduced protein is expressed in excess this may lead to changes in cellular behaviour.
With the issues inherent with these various approaches, you have to ask: “Is there a better way?”
1. Skogs M Antibody validation in bioimaging applications based on endogenous expression of tagged proteins. J Proteome Res. 2016 Oct 11