Volume 35, No. 1, January 1960
Printed in U.S.A.


Tamir K. Nassar, Marietta Issidorides and William M. Shanklin

Department of Histology, School of Medicine, American University
of Beirut, Beirut, Lebanon

Received for publication July 7, 1959

ABSTRACT.- Representative pieces of human brain were fixed in 10% formalin, embedded in paraffin and sectioned at 5 µ. Paired sections were used, one of which was oxidized in equal parts of 03% potassium permanganate and 03% sulfuric acid for 1-2 min, while the other was left unoxidized. Both the oxidized and unoxiducd sections were impregnated with silver diamine. The lipofuscin granules in the nerve cells appeared as small intensly stained black dots, surrounded by a clear unstained zone, in the unoxidized sections, while in the oxidized sections there was an outer ring of intensely blackened material surrounding a central unstained dot.


In an earlier paper we (D'Angelo. Issidorides and Shanklin 1956) stained lipofuscin granules in human neurons with fuchsin paraldehyde, Gomori's chrome hernatoxylin. Sudan Black B, carbol fuchsin and periodic-Schiff methods. The individual lipofuscin granules in many areas of the brain reacted to these methods in ways which led us to conclude that each granule is built of several concentric layers which react differently to the several staining methods applied.

We have investigated this problem fturther with the purpose of developing a technic which would more clearly reveal the layered structure of the lipofuscin granules. In our current studies (Sihinklin and Isssdoridess l959) of nerve lipofuscin we have used histochemical methods which suggested the presence of free amino and carboxyl groups, cystine, unsaturated fats, protein-bound sulf-hydryl and disulfide groups, and thereby supported the view that these granules represent either a lipoprotein (Einarson 1953) or proteolipid complex (Dixon 1954). We believe, however, that the internal structure of the lipofuscin granules is also of primary importance in establishing their physiological role, if any, in the neuron.

We have found that a modification of previous silver impregnation methods developed for paraffin-embedded material (Nassar and Shanklin 1951) was very selective for the demonstration of lipofuscin granules (Shanklin, Issidorides and Nassar 1957).

In this study we have used our (Shanklin. Issidorides and Nassar 1957) silver impregnation method combined with oxidation by acidified potassium permanganate.


Our staining and oxidation procdur is as follows:

  1. Fix small pieces of fresh brain tissue in 10% neutral formalin for 1 wk.
  2. Wash in tap water, dehydrate, clear in xylene and embed in paraffin.
  3. Cut sections at 5µ.
  4. Deparaffinize and carry paired sections through absolute to 95% alcohol. Carry one section through steps 5 to 10; take the other section directly to step 11 below.
  5. Hydrate sections and wash in distilled water.
  6. Oxidize sections in equal parts of 0.5% potassium permanganate and O.5% sulfuric acid for 1-2 min until sections turn a brownish color. The solution must be changed frequently.
  7. Rinse in distilled water.
  8. Decolorize in 2% aqueous oxalic acid for 2 mm.
  9. Wash under tap water for 5 min.
  10. Pass sections through graded alcohols to 95%.
  11. Place both oxidized and unoxidized sections in 2% silver nitrate to which pyridine 3 drops/10 ml have been added, for 4-5 hr at 50-55° C.
  12. Impregnate in silver diammino-hydroxide solution containing pyridine 3 drops/ 10 ml, for 5 mm at 50-55° C.
  13. Rinse quickly in 95% alcohol.
  14. Reduce in an equal-parts mixture of 2% neutral formalin and absolute alcohol for 2 min.
  15. Wash in distilled water.
  16. Tone in 0.2% aqueous gold chloride until the sections turn greyish.
  17. Fix in 5% sodium thiosuiphate for 2 min.
  18. Wash in tap water and counterstain lightly in eosin or 1% aqueous erythrosin.
  19. Dehydrate, clear and mount.

All sections were impregnated with silver, half following oxidation with potassium permanganate and sulfuric acid and half without oxidation. X 5OO.
Fig. 1. Nerve cell in thalamus, unoxidized, silver impregnation. Arrow indicates clear unstained zone of granule.
Fig. 2. Nerve cell in thalamus, oxidized, silver. inpregnation.
Fig. 3. Cell in nucleus of the superior colliculus, unoxidized, silver impregnation.
Fig. 4. Cell in nucleus of the superior colliculus, oxidized, silver impregnation


In all sections impregnated with silver diammine without previous oxidation (figures 1, 3, 5 and 7). the lipofuscin granules appear as small distinct black dots surrounded b a clear zone. In all sections impregnated with silver diammine after initial oxidation (figures 2, 4, 6 and 8) the granules appear as black rings having a clear center. This is best observed in those parts of the cells where the granules are fewer and not overlapping. The diameter of the rings is approximately twice that of the dots.

Fig. 5. Reticular cell in the medulla,. unoxidized, silver impregnation.
Fig. 6. Reticular cell in the medulla, oxidized, silver impregnation.
Fig. 7. Cell in arcuate nucleus, unoxidized, silver impregnation
Fig. 8. Cell in arcuate nucleus, oxidized, silver impregnation.

From our figures it is implied that two layers of substances, concentrically arranged, are present in each granule. With our silver impregnation, in the absence of oxidation, silver deposition occurs on the inner layer, or core, of the granule. The outer layer is entirely unstained, but its presence is well inferred, especially in figure 1 (arrow). When we introduce the oxidative step, the substance of the inner layer loses its ability to accumulate silver deposits, while the substance of the outer layer is well impregnated. At this stage it is difficult to interpret our results in terms of chemical events since formalin is present in both alternative procedures as the reducer the reducer. We tried our method, however, omitting formalin and obtained the following results. In the unoxidized sections the center, or inner layer, of the granules reduced silver intensely, while the outer layer remained clear. In the oxidized sections reduction of silver occurrcd agin on the inner layer, but to a lesser degree, while the outer layer remained clear. Further studies are in progress on this aspect of the problem.

It appears then that for the demonstration of the outer layer of the lipofuscin granule both formalin and the KMnO4-oxalic sequence are necessary, i.e. this outer layer becomes argyrophil (not argetaffin) only after oxydation.


This work was supported by a grant (B-1701) from the National Institutes of Health. United States Public Health Service.


D'ANGELO, C., ISSIDORIDES, M., and SHANKLIN, W.M. 1956. A comparative study of the staining reactions of granules in the human neuron.. J. Comp. Neur. 106, 487-506.

DIXON, K.C., 1954. Cytochemistry of cerebral grey matter. Quart. J. Exp. Physiol., 29, 129-51.

EINARSON, L. 1953. Deposits of fluorescent acid fast products in the nervous system and skeletal muscles of adult rats with chronic vitamin E deficiency. J. Neur. Neurosurg. Psychiat. 16, 98-109.

NASSAR, T.K., and SHANKLIN, W.M. 1951. Staining neuroglia with silver diammino-hydroxide after sensitizing with sodium sulfite and embedding in parrafin. Stain Techn., 26, 13-18.

SHANKLIN. W. M., and ISSIDORIDES, M. 1959. A preliminary report on the histology and histochemistry of the human cerebellum, hypothalmus and medulla. Abstract of Paper Read from Platform, Summer Meeting 1959. Anatomical Society of Great Britain and Ireland.

SHANKLIN. W. M., ISSIDORIDES, M. and NASSAR, T.K. 1957. Neurosecretion in the human cerebellum. J. Comp. Neur., 107, 315-38.

1Research supported by a grant (B-1701) from the National Institutes of Health. United States Public Health Service.

The above article is transcribed from a copy at the AUB's Saab Medical Library.


© AUB Libraries, 2005

Last modified: Sun Oct 8 12:45:00 2006 BL