Differentiation is the process of removing excess dye from tissues in order to accentuate a structure which retains the dye while all about are losing theirs. It is similar to decolorising, but infers a high degree of selectivity. The difference between differentiation and decolorising is that decolorising is non-selective and removes almost all of the dye from a section non selectively. Apart from this, the two processes are essentially the same.
There are a limited number of ways that differentiation can be accomplished. These are:–
When we are using a simple solution of a dye to stain the tissue, it is usually feasible to remove excess dye from the tisuue by washing in the solvent. At its simplest this is seen with aqueous eosin Y staining in an H&E. It is customary to lightly differentiate the eosin by washing with tap water for a short time to remove excess dye and get the differential effect of which eosin Y is capable. A second example would be found in fat staining with a dye such as oil red O, which is often differentiated with 70% ethanol, a standard solvent for that dye.
Solvent differentiation works because of the nature of solvents. In the case of eosin a polar solvent, water, is used. Polar solvents have an attraction for compounds that can ionise and tends to dissolve them. The charged ions of the dye are attracted to the complementary charge on the appropriate side of the water molecule and are removed from the tissue. Of course, this is far from instantaneous. The dye is bound fairly strongly to the tissue, and this must be overcome before the dye can dissolve in the solvent. At any given time only a few molecules will be removed, so it is a slow process compared to some others. When extended, however, it can all but decolorise a section. In addition, water, in particular, often has significant amounts of dissolved substances. These can impact on the destaining process, often because they change the pH (see below).
Non-polar solvents can also remove material from the tissue. In the case of oil red O mentioned earlier, the dye can dissolve into the solvent (70% ethanol). Since lysochromes stain by preferential solubility, it would be expected that application of a suitable solvent would extract some dye simply by virtue of being a solvent.
The effectiveness of solvent differentiation depends to a large effect on the strength with which the dye is bound to the tissue, the type of bond involved, the degree of attraction between the solvent and the dye, the volume of the solvent and the time for which it is applied.
This is really just a fancy way of saying dilute acids and bases, and is probably the commonest approach to differentiation. It is worth stressing that differentiation with dilute acid is not the only way pH differentiation can be done. Dilute alkalis (bases) can also be used, as can buffer solutions. The acids are used to remove basic dyes or lakes, and dilute bases remove acid dyes.
The action of acids in this context is very similar to their effect as accentuators, but in reverse. An acid pH simultaneously causes an increase in the ionisation of amino groups and a reduction in the ionisation of carboxyl and hydroxyl groups. This applies to both tissue groups and those on the dye. In effect, the acid used to differentiate will compete with the anions of the tissue for the dye cations. Since the differentiating acid is deliberately chosen to be a stronger anion than the tissue groups, it will dissociate more thoroughly and will invariably win the competition. The result is that the dye is preferentially removed from the tissue by dissolving it out.
Bases work much the same with acid dyes. They compete with tissue cations for the anions of dyes and disrupt their attachment to the tissue. The dye is removed by dissolving into the differentiating solution.
Controlling this process is quite simple. The rapidity with which the dye is removed can be adjusted by changing the concentration of the acid (or base) on the principle of weaker solutions taking longer. Alternatively, a weaker acid that dissociates less could be substituted. Likewise, a less polar solvent will tend to slow down the differentiation due to lower ionisation of the acid. Finally, when enough dye has been removed, simply remove the acid and replace with a solvent. However, time and concentration are the two most important factors.
Hydrochloric acid at 1% in 70% ethanol is likely the single most used solution of this kind, primarily because of its importance in differentiating regressive alum hematoxylin stains. Baker showed that in the case of mordant dyeing, such as with hematoxylin, the acid disrupts the tissue - mordant bond rather than the mordant - dye bond.
Although buffers could be used for this purpose, they very rarely are. The process is so simple to control that precise pH adjustment is of no real benefit in practice.
In those cases where a mordant has been used in conjunction with a dye to stain the tissues, the mordant may also be used to remove excess dye. This is accomplished by applying a solution of the mordant to the stained section. The mordant in solution will have the same affinity for the dye as the mordant attached to the tissue has. In this way, the excess mordant in solution competes with mordant attached to the tissue for the limited amount of dye. Since the mordant in solution is in considerable excess, it wins the competition and slowly removes the dye from the tissue. It is possible to completely remove all dye from sections in this manner. The classic example of this kind of differentiation is Heidenhain's iron hematoxylin.
In contrast to differentiation of lakes with acids, which disrupt the tissue - mordant bond, mordant differentiation leaves the mordant attached to the tissue. Only the dye is removed. Thus sections can be restained by immersing them in a solution of the dye only.
The process can be controlled by adjusting the concentration of the mordant, and limiting the time for which it is applied. The disadvantage of this technique is that some mordant - dye combinations produce unsharp results. A case in point would be using potassium alum solutions to differentiate alum hematoxylin stains. This is accomplished far more easily with acid alcohol, and the results are sharper and more clearly defined. Although Heidenhain's iron hematoxylin is traditionally differentiated in this manner, it too can be done with acid alcohol more rapidly.
Most dyes lose their colour when oxidised (bleached). Although it is therefore possible to differentiate stained sections with oxidising agents, it is rarely used in practice because most bleaches are far too strong and difficult to control, producing uneven results. Quite mild oxidising agents are required, but these may take quite a bit longer than other methods of differentiation.
The classic example is found in the demonstration of myelin using the Weigert - Pal technique and its variations. After staining, differentiation is accomplished with a potassium permanganate - oxalic acid bleach, and/or a borax ferricyanide bleach, depending on the particular variation used.
This is sometimes called displacement and refers to processes where one dye replaces (or displaces) another. It is based on the principle that compounds with similar structures can substitute for each other. Where the similarity is the carboxyl or amino groups of the dye, then similarly charged dyes can replace one another.
If an acid dye is used to stain tissue, then is washed off and a contrasting colour acid dye is applied, it will slowly replace the first dye. This is the underlying principle of many staining methods, in particular the various sequential trichrome methods.
Baker, John R., (1958)
Principles of biological microtechnique
Methuen, London, UK.