Schiff's reagent is a very sensitive means of detecting aldehydes, and can be used in a method to demonstrate deoxyribosenucleic acid (DNA) specifically, in contrast to unstained ribosenucleic acid (RNA). This method is the nucleal reaction of Feulgen and Rossenbeck, usually simply called the Feulgen stain or reaction. It is usually done with pararosaniline Schiff solution, but it works well with some others, including the fluorescent acriflavine solution.
The technique involves treating sections with dilute hydrochloric acid. This treatment hydrolysis DNA and removes the bases. The sugar remains and reacts as an aldehyde, including condensation with Schiff's reagent when it is subsequently applied. The typical red colouration where DNA was present is thus formed. The diagrams below illustrate the removal of the bases and the subsequent availability of an aldehyde, with the relevant parts in red.
|+ H2O =|
|Deoxyribose with base||Deoxyribose hemiacetal||Deoxyribose aldehyde|
The nucleal reaction is considered to be specific for DNA. The other nucleic acid, RNA, does not react the same way either because the acid hydrolysis causes it to dissolve away, or because of the hydroxyl group present in ribose which has lost its oxygen in deoxyribose. It is important to note, however, that if acid hydrolysis is applied for too long, especially at elevated temperature, then DNA also can be completely removed, and this is a known source of failure in the technique.
It would seem that two things happen during acid hydrolysis of DNA. The first is that the purine bases are removed and aldehyde groups on deoxyribose are formed. The second is that histones (protein associated with DNA) and apurinic acids (deoxyribose without the base) are progressively removed. Hydrolysis should be stopped and Schiff's reagent applied when the first is well on its way, but before the second has removed too much of the aldehyde.
If there is any doubt about the nucleal reaction's specificity, then parallel sections should accompany the test sections with the acid hydrolysis ommitted. Any red colouration in the unhydrolysed section should be subtracted while evaluating the test section. If both sections are unstained, or unsatisfactorily so, then either the Schiff's reagent is no longer usable or, more likely, the hydrolysis has been applied for too long.
In the original method the hydrolysis is done with 1N-hydrochloric acid at 60°C for an appropriate length of time, varying according to the fixative used. Strongly acid fixatives should be avoided since they may hydrolyse DNA during fixation, perhaps to the extent that all DNA is removed and there is no staining. Bouin's picric acid, acetic acid, formalin mixture is known to do this, although sometimes DNA can be demonstrated if the acid hydrolysis step is ommitted during staining. Regardless of the fixative used, it is recommended that trials be conducted so the optimal hydrolysis time can be determined. The following may act as a guide:
|Formalin||8 minutes||NBF, formal saline, etc.|
|Formaldehyde vapour||30-60 minutes||On freeze dried tissue.|
|Formal sublimate||8 minutes||B5 has a very similar formula.|
|Zenker-formal||5 minutes||Zenker-formal and Helly are different.|
|Bouin||Not recommended||Refers to Bouin's picric-acetic-formalin.|
|Glutaraldehyde||Not recommended||Precede hydrolysis with an aldehyde block.|
It has been shown that the 60°C temperature used in the original 1N-hydrochloric acid hydrolysis can result in loss of stainable material, and it is recommended that 5N-hydrochloric acid at room temperature for 30 minutes or longer be used. This allows for less precise timing and darker staining. Pearse gives some times as a guide:
|Alcoholic fixatives||20 minutes to 2 hours|
|Formalin fixatives||35 minutes to 4 hours|
|Formaldehyde vapour||2 hours to 8 hours||Freeze dried tissue|
Pearse, A. G. E., (1968, 1972)
Histochemistry: Theoretical and Applied, Ed. 3
Churchill Livingstone, Edinburgh, London, UK
Kiernan. J.A., (1999)
Histological and histochemical methods: Theory and practice, Ed. 3
Butterworth Heinemann, Oxford, UK.