115

First Discussion Period *

DE ESTABLE: One of the virtues of this Symposium, whose organizers deserve the
most sincere praise, consists in its fostering new research.
I feel that I am expressing a unanimous judgement when I say that all the reports
we have listened to are excellent. This statement does not mean that a unanimous
agreement should exist on the many problems here debated.
It is necessary that the complex unity of the concepts obtained by integration should
prevail upon the simple and schematic unity obtained by elimination.
The prevailing characteristic of the physiology of the nervous system was inter-
preted as an exclusive function of the neurons and it was even claimed by a great
physiologist (Nobel laureate) that the axon stands in relation to Neurology as the
atom stands in relation to Chemistry and the electron to Physics!
It is true that, starting with the ideas of Cajal, Achticarro and del Rio Hortega, it
was thought that the neuroglia would play an important role in the process of myelin-
ation, in the isolated propagation of the excitatory status, in the trophism of the
neuron and it was even speculated that it could secrete some hormone (at present,
Bremer is of the opinion that perhaps the neuroglia could elaborate pseudocholin-
esterase).
On the basis of outstanding experiments with microelectrodes, microdissection
techniques, ultramicrocytochemistry and microscopic and submicroscopic control,
much progress has been made in the knowledge of the meaning of the neuroglia in
some aspects of the physiology of the nervous system and its physiopathology (such
as the importance of the neuroglia in the blood-brain barrier and in brain oedema).
We have here among us some of the people who are to be credited for this advance-
ment. Without naming them, it is well known who they are.
One has the tendency to conceive the functions of the neuroglia and the neuron as
intimately related. With a certain amount of exaggeration, the importance of the first
has been magnified, lessening the importance that the classic investigators and the
majority of the modern ones ascribe to the second. It would be unproper for the
scientific mind to choose sides and become appassionated for any other thing but
truth.
Four factors should be kept in mind in considering the nervous system life (function-
ality). These are: (I) the neuron and its synaptic spectrum; (2) the glia cells, their
contaets between them and with: (a) the neurons, (b) the vessels, (capillaries, arteries,

* This discussion refers to the papers of Prof. E. D. P. De Robertis, Dr. M. Polak, Dr. R. L. Friede,
Dr. A. Lasansky, Dr. I. Klatzo et al., Dr. J. Miquel and Dr. W. Haymaker.
References p. I23
116 FIRST D I S C U S S I O N P E R I O D

veins); (3) the blood vessels; (4) the molecules flowing in the narrow intracellular
spaces (internal media) and the molecules which are not ordered in structures or
substructures, and cannot be detected either with the electron or the light microscope.
The metabolism of the neuron does not depend solely on the glia activity; when the
synaptic spectrum is destroyed without damaging the neuroglia, the neuron undergoes
atrophy and dies.
On the other hand every glia cell (even astrocytes) is not in contact by endfeet with
capillaries. Millions of them are only related with arteries and veins. It could be
inferred, therefore, that the neuroglia has the substance flowing in the intracellular
space as a nutrient medium. The nerve cell contacts although rarely with the capillary

Fig. 1. Ventricular ependyma of a rat irradiated in the head with an X-ray load of 20,000 roentgen.
Osmic perfusion fixation 24 h after irradiation. PAS and hematoxylin staining following dimedone.
The arrow shows the glycogen evenly distributed in an ependymary cell.
Fig. 2. Cerebellar cortex of rat fixed by osmic perfusion 24 h after receiving an X-ray load of 20,000
roentgen. PAS and hematoxylin staining after dimedone. Notice the positive PAS material in the
perikaryon of the Von Bergmann cells (BC) and their processes.
 FIRST DISCUSS ION P E R I O D 117

and it should be emphasized that the cytoarchitecture is modelled in accordance with

the angioarchitecture and not with the glia architecture.
Undoubtedly, each glial type and particularly the astrocytes, perform multiple
functions. That the astrocytes participate in the circulatory regulation of the nervous
system mitigating the effect of the beating vessels on neurons is inferred from ex-
periments made with vasodilators and vasoconstrictor drugs and with experimental
oedema provoked by substitution of blood with perfusion liquids. The existence of
physiological sphincters along the vessels related to the glia cells corroborates this
hypothesis.
It is not possible in this short lapse of time, to compare the innervation of the vascu-
lar system of the neuroaxis with the one existing along the vessels of other organs. It
is surprising that in the nervous system the vascular innervation is poorer than in
other systems. The glia cells may compensate this deficiency.
Some confusion exists among some functional types of glia with neurons, some
neurons are interpreted as glia. This mistake can be opposed to that made by investi-
gators of the last century who described glia cells as being neurons. The differential
diagnosis cannot be discussed here but the point seems clear to most researchers.

ESTABLE-PUIG: We would like to show here some light and electronic photomicro-
graphs of our irradiation material. We have studied this (De Estable et al., 1965) with
a modification of Palay’s osmic perfusion fixation method (Palay et al., 1961) and we
have attempted to establish if this technique could be useful for parallel optic and 3

electronic microscopy studies in order to do what could be called ultrastructural

histochemistry, particularly with glycogen (De Estable et al., 1963). Previous investi-
gations have demonstrated the appearance of glycogen following ionizing radiation
(Klatzo et al., 1961; Miquel et al., 1963; Kruger and Maxwell, 1963). This material
corresponds to X-ray irradiated rat brains perfused with an osmic solution. From

Fig. 3. Similar to the previous Fig., with greater enlargement. Notice that the glycogen is restricted
to the glial cells and is not seen in the cytoplasm of the Purkinje cells. Many grains with pyknotic
nuclei can be observed. BC = Von Bergmann cell.
References p . 123
118 FIRST DISCUSSION PERIOD

those brains blocks for electronic-microscopylwere imbedded in Epon and the re-
maining material was imbedded in paraffin for light microscopy. Following removal
of paraffin the sections were treated with dimedonemfor the practically; specific
differential block of the nonglycogenic aldehydic groups, following Bulmer’s technique
(Bulmer, 1959). With this technique the PAS staining shows the glycogen without the
disturbing background staining which results from osmic fixation of the nervous
tissue. The specificity of such method was demonstrated by amylatic digestion,

Fig. 4. Cerebellar cortex of rat fixed by osmic perfusion 24 h after administration of an X-ray load
of 20,000 roentgen. Topographic photomicrography of the limiting zone between the myelinic
axis of the cerebellar lamellae and the granular layer. Notice that the tissue appears very compact,
without the usual artifacts of fixation. Numerous myelinic fibers (FM) sectioned at different angles
and glial fibers can be seen. Epon embedding. Lead staining following Karnovky method. x 17,000.

Our observations have indicated that this is an excellent method to preserve and
demonstrate different cellular elements and particularly glycogen. The latter appears
evenly distributed in the cytoplasm without the heavy artifactual precipitates that
follow picroalcoholic fixation.
The first figure corresponds to the periventricular cerebral tissue. Blood vessels are
 FIRST DISCUSSION PERIOD 119

dilated and empty on account of the perfusion. Some ependymary cells contain
glycogen evenly distributed in the cytoplasm.
Fig. 2 shows the cerebellar cortex of a rat which received 20,000 roentgens 24 h
before being dispatched. Hardening and staining was similar to that of the previous

Fig. 5. Cerebellar cortex of rat with osmic perfusion fixation 24 h after an X-ray load of 20,000
roentgen in the head. Electronic photomicrography of the granular layer in which morphologically
preserved cells can be seen. The arrows show the modified grains. Epon embedding. Lead staining
following Karnovky method. x 13,000.

case. We believe that this suggests that the PAS-positive staining of the molecular
layer corresponds with the outline of the Von Bergmann cells. Glycogen is mainly
located in these cells but not in the Purkinje cells. A small amount of glycogen can be
found in some astrocytes of the granular layer close to the Purkinje cells.
Many pyknotic cells can be seen in the granular layer. This is a well-known fact to
those studying the radiobiology of the nervous system, since granular cells are the
neuronal elements most sensitive to ionizing radiation injury. We do not know yet
References p . 123
120 FIRST DISCUSSION PERIOD

which is the relation of these neurons, which are known to have large branches
ramified in the molecular layer, with the glycogen accumulation in the glial cells. Fig. 3
shows with greater enlargement one area of the previous field. Pyknotic granular cells
are better seen and the glycogen accumulation restricted to the glial cells can also
be clearly observed.

Fig. 6. Similar to the previous figure showing a pyknotic granule and an astrocyticprocess filled with
glycogen. Osmic perfusion fixation. Epon embedding. Lead staining following Karnovky method.
x 60,000. G1= glycogen; PA = astrocytic process.

Fig. 4 is a topographic electronic photomicrography showing the general appear-

ance of the limiting zone between the myelinic axis of the cerebellar lamellae and the
granular layer as seen when the osmic perfusion fixation method is employed, the
tissue appearing quite compact. The endothelium of the empty vessels dilated by the
perfusion can be observed in the lower right angle of the figure. Many myelinic fibers
sectioned at different angles can be seen exhibiting neurofilaments very clearly. Glial
fibers are seen across the field in the upper part of the Figure.
 F I R S T DISCUSSION P E R I O D 121

Fig. 5 is an electronic photomicrography of the granular layer where the morpho-

logy of some normal grains can be observed with the large nucleus and its nucleole
and the scarce cytoplasm withmitochondria and RNA particles, which are the function-
al equivalent of Nissl substance in other neurons. Some granular cells are abnormal

Fig. 7. Rat cerebellar cortex with osmic perfusion fixation 24 h after receiving an X-ray load of 20,000
roentgen. Electronic photomicrography of an area of the molecular layer showing a Von Bergmann
cell process filled with glycogen granules. Epon embedding. Lead staining following Karnovky
method. x 60,000. PA = astrocytic process.

showing a reduction of their diameter as a consequence of the retraction of the

cytoplasm and particularly of a retraction and densification of the nucleus, where
lacunar spaces with granular content can be seen. With greater enlargement Fig. 6
shows that the glial processes adjacent to the pyknotic cells are completely filled with
grains stained with Karnovky’s technique. These grains are interpreted as glycogen
inclusions on account of their size, the lead affinity and the correlation with obser-
vations performed with light microscopy.
References p. 123
122 FIRST D I S C U S S I O N P E R I O D

The Von Bergmann cell processes in the molecular layer are also filled with glycogen
granules which are also found in the astrocytic processes of the neuropil but not in
the nervous fibers, nor in the axonal ends of those dendrites making synaptic contact
there (Fig. 7).
The astrocytic nature of the space where the glycogen granules are found is certified
by the fact that they are found coexisting with glial fibers (Fig. 8).

Fig. 8. Similar to the previous Fig. showing the coexistence of glycogen grains and glial fibers (FG)
on the left upper angle. Lead staining. Epon embedding. GI = glycogen; PA = astrocytic process.

At present we are performing a serial study with the purpose of making certain the
glycogen formation mechanism and its possibilities of relation with different cellular
organs.

Luco: According to the law of Langley and Anderson, it seemed impossible to

re-innervate with cholinergic fibers the smooth muscles that are normally innervated
by adreneIgic fibers. In a paper reported by Vera, Vial and Luco, functional inner-
 FIRST DISCUSSION PERIOD 123

vation of the nictitating membrane of cats was obtained by hypoglossal nerve. This
heterogeneous innervation was demonstrated by the contraction of the smooth muscles
during the stimulation of the hypoglossal nerve and by the normalization of the
sensitivity to adrenaline, noradrenaline and acetylcholine.
The myelinated fibers of the hypoglossal nerve grew through neuron tubes of
non-myelinated fibers of the postganglionic sympathetic nerve. The observation of
Simpson and Young was confirmed i.e. the growing fibers appear at the periphery as
myelinated fibers. But the terminal portion of the fibers in our experiments was different
from those observed in the hypoglossal nerve that innervate the tongue muscles. The
terminals in our experimental situation were very much like the normal terminals of
the sympathetic system in the nictitating membrane.
It was suggested that the Schwann cells are not able to wrap the nerve terminals
because a peripheral factor from the smooth muscles is blocking this reaction.
Furthermore, it can be thought that in order to activate the fibers of the smooth
muscles the presence is necessary of a long nude terminal to permit the release of a
relatively large amount of chemical mediator, that may reach all the muscle fibers.
These histological observations under the optical microscope suggested a problem
on the relation between the axon and the reaction of the Schwann cells in wrapping
theaxon. In other words, is this reaction induced by the axon or are there specific
Schwann cells that result in myelinated or non-myelinated fibers?
The electron microscopic observations recently made by Dr. Vial in the preparation
of hypoglossal nerve-nictitating membrane confirmed the previous observations and
added new data. At the periphery, the myelin is formed by ‘tongue’-like lamellae
at the outer layers of the myelin, a situation not seen at a normal myelinated motor
nerve.

REFERENCES

BULMER, D., (1959); Dimedone as an aldehyde blocking reagent to facilitate the histochemical
demonstration of Glycogen. Stain Technol., 34, 95-98.
DE ESTABLE, R, F., ESTABLE-PUIG, J. F., AND HAYMAKER, W., (1965); Electron microscopy of cere-
bellar cortex following ionizing radiation injury. Znt. J. Neurol. (Montevideo), in press.
DE ESTABLE, R. F., ESTABLE-PUIG, J. F., AND MIQUEL,J., (1963); Fixation and identification of
glycogen in the nervous system for light and electron microscopy studies. 39th Meeting of the
American Association of Neuropathologists.
KLATZO,I., MIQUEL,J., TOBIAS, C., AND HAYMAKER, W., (1961); Effects of alpha particle radiation
on the rat brain including vascular permeability and glucogen studies. J. Neuropath. exp. Neurol.,
20,459483.
KRUGER,L., AND MAXWELL, D., (1963); Electron microscopy of radiation induced laminar lesions
in the cerebral cortex of the rat. The Second International Symposium on the Response of the
Nervous System fo Ionizing Radiation. University of California, Los Angeles, August 29-31.
MIQUEL,J., KLATZO, I., MENZEL, D. B., AND HAYMAKER, W., (1963); Glycogen changes in X-irradia-
ted brain. Acta neuropath. (Bed.), 2, 482490.
PALAY,S. L.; MCGEERUSSELL, S. M., GORDON, S., AND GRILLO,M. A., (1961); Fixation of neural
tissue for electron microscopy by perfusion with solutions of osmium tetroxide. J. Cell Biol., 12,
385.