Electron Microscopy of Axon Degeneration: A Valuable Tool in Experimental Neuroanatomy

33 Pages·1966·2.496 MB·English

by  Dr. John F. Alksne, Dozent Dr. Theodor W. Blackstad, Prof. Dr. Fred Walberg, Dr. Lowell E. White Jr. (auth.)

Ergebnisse der Anatomie und Entwicklungsgeschichte
Reviews of Anatomy, Embryology and Cell Biology
Revues d'anatomie et de morphologie experimentale
Herausgegeben von
A. Brodal, Oslo . O. Elze, München· W. Hild, Galveston . R.Ortmann, Köln
G. Töndury, Zürich· E. Wolft, Paris
Schriftleitung
G. Töndury, Zürich
Band 39 . Heft 1
lohn F. Alksne, Theodor W. Blackstad, Fred Walberg
and Lowell E. White jr.
Electron Microscopy of Axon Degeneration:
A ValuableTool in
Experimental Neuroanatomy
W ith 20 Figures
Springer-Verlag Berlin Heidelberg GmbH 1966
Dozent Dr. Theodor W. Blackstad, Prof. Dr. Fred Walberg, Anatomisk Institutt,
Universitetet i Oslo, Oslo 1JNorge, Karl Johans gate 47 (Domus Media)
Dr. John F. Alksne, Harbor General Hospital,
1000 W. Carson Btreet, TorranceJCalif. 90509, U BA
Dr. Lowell E. White jr., The Medical School, University of Washington,
Beattle 5, W ash.JU BA
Supported by Grants NB-02215 and NB-02896 from the National Institute of Neurological Diseases and
Blindness, U. S. Public Health Service. This aid is gratefully acknowledged. For technical assistance the
authors are indebted to Mrs. J. Line Vaaland, Mr. B. V. Johansen and Mr. E. Risnes
ISBN 978-3-540-03493-3 ISBN 978-3-662-30450-1 (eBook)
DOI 10.1007/978-3-662-30450-1
Alle Rechte, insbesondere das der Übersetzung in fremde Sprachen, vorbehalten. Ohne ausdrückliche
Genehmigung des Verlages ist es auch nicht gestattet, dieses Buch oder Teile daraus auf photomechanischem
Wege (Photokopie, Mikrokopie) oder auf andere Art zu vervielfältigen
© by Springer-Verlag BerlinHeidelberg 1966. Library of Congress Catalog Card Number 64-20582
Originally published by Springer-Verlag Berlin Heidelberg New York in 1966.
Titel-Nr. 4453.
Die Wiedergabe von Gebrauchsnamen, Handelsnamen, Warenbezeichnungen nsw. in diesem Werk berechtigt
auch ohne besondere Kennzeichnung nicht zu der Annahme, daß solche Namen im Sinn der Warenzeichen
und Markenschutz-Gesetzgebung als frei zu betrachten wären und daher von jedermann benutzt werden dürften
Contents
Page
1. Introduction . . . . 6
II. Material and methods 7
III. Observations and comments 13
l. The hippocampal region. 14
a) Molecular layer of area dentata 14
b) Layer II of presubiculum . . . 16
c) Molecular layer of subiculum and hippocampus 16
d) Pyramidal and adjacent layers in hippocampus (regio superior) 19
2. The brain stem. . . . . 20
a) The inferior olive . . 20
b) Dorsal column nuclei . 23
IV. Discussion . . . . . . . . 25
a) Survival time 25
b) Appearance of degeneration. 27
c) Relation to silver impregnation 28
V. Summary 29
Bibliography 29
Index ... 32
I. Introduction
Experimental methods for the mapping of nervous pathways are based
partlyon the study of retrograde processes in the perikaryon, partlyon the
demonstration of degenerative processes along the peripheral part of a transected
axon. For this purpose, the Marchi method by which a selective staining of
degenerating myelin is obtained has been extensively used. However, when this
method is used the non-myelinated terminals of the transected axons are not
stained. The introduction, about two decades ago, of silver impregnation as a
means of tracing degenerating axons (especially the Glees and Nauta methods)
by which also terminal boutons can be demonstrated, led therefore to revolutionary
progress in the investigation of interneuronal connections. Notwithstanding, there
are weH known difficulties involved in this kind of research. The capriciousness of
the silver methods not seldom results in failure of impregnation with loss of
valuable experimental animals. But even when well impregnated sections are
used, other fundamental difficulties exist. One of the major problems is to prove
beyond doubt that the impregnated structures are degenerating boutons and not
merely fragments of non-terminal fibres passing the area under examination.
Furthermore, only on occasion will silver impregnation permit one to accurately
define the specific part of the receiving neuron on which the impregnated fibres
end, i.e., whether the bouton makes contact with soma, dendrite or spine. This
is partly due to the fact such structures are usuaHy poorly defined in reduced
silver preparations, partly because submicroscopic tissue elements, particularly
astroglial sheets, may separate structures appearing in the light microscope to be
in immediate contact. This latter fact has become clear through electron micro
scopy. Finally, distinction between the different kinds of membrane specialization
at synaptic contacts which have been demonstrated by electron microscopy is
entirely beyond the capabilities of light microscopy.
The aim of the present paper is to present the neuroanatomist with examples
of the practical use of electron microscopy for mapping the exact termination
of central nervous pathways by means of studying the distribution of degenerating
boutons. Separate papers have appeared or will appear on most of the connections
here only briefly dealt with. It is not our purpose to present new cytological data.
The first detailed account on the changes in the fine structure of terminal
boutons following interruption of their parent fibres was given by GRAY and
HAMLYN (1962) in their studies of the optic tectum in the chick. The features
of terminal axonal degeneration in the cerebral cortex of mammals were later
clarified by COLONNIER and GRAY (1962) and by COLONNIER (1964). The boutons
belonging to transected axons were found to shrink and darken and become
engulfed in glial cytoplasm together with the attached postsynaptic element, the
spine. With increasing survival time after operation the pre- and postsynaptic
elements experienced further loss of their normal fine structure. Within some of
the regions studied by us, essentially the same phenomena have been encountered.
J. F. ALKSNE et al.: Electron microscopy ofaxon degeneration 7
In others, the findings are düferent, showing that regional differences exist. This
paper will support Colonnier's conclusion that "the identification of degenerating
endings by electron microscopy may be a valuable adjunct for the determination
of the precise ending of fibre tracts in the central nervous system, especially when
the Nauta or Glees methods are unsuitable".
11. Material and methods
In our studies cats and rats were used. Over the years a variety of preparatory
methods have been employed. The demonstration ofaxonal degeneration has not
proved to depend critically on the methods of fixation and embedding beyond
that of securing a general good preservation of the tissue. In part, fixation has
been by perfusion, with 10% formalin containing a varying amount of sucrose and
buffered with phosphate, the vascular system first having been briefly rinsed
with Ringer's solution. In most cases, the formalin fixative introduced by HOLT
and HICKs (1961) has been used. The procedure followed in the cat has been
described elsewhere (WALBERG 1963a). In the rat, perfusion through the abdominal
aorta has been found to be advantageous. This, in contrast to an approach through
the thoracic aorta or the left ventricle, does not interfere with respiration up to
the moment of penetration of the fixative. Probably, perfusion with a glutar
aldehyde fixative (SABATINI et al. 1963) will prove as adequate for degeneration
studies as perfusion with formalin. In part we have fixed the tissues by direct
immersion in 2 % osmium tetroxide solution, essentially according to P ALADE
(1952), CAULFIELD (1957) or MILLONIG (1962). Material from animals perfused
with formaldehyde was always post-fixed in one of these solutions. The fixatives
and the dehydrating media were used chilled. As an embedding material,
methacrylate (butyl/methyl 9: 1) was used in our earlier studies, but later it was
replaced by Araldite (WEBSTER et al. 1961).
In neuroanatomical work of the kind here under consideration, fuH under
standing of topography and complete identification of the structures under
examination is essential. Great care should, therefore, be taken at every step of
dissection, trimming and microscopy that these requirements are fulfilled (Fig. 1).
On the cut surface of a brain fixed by perfusion and removed from the cranium,
the separate nuclei and fibre tracts can be discerned under the dissecting microscope
and isolated with fragments of razor blades. The architectonic details become
particularly conspicuous in osmium tetroxide fixatives, used alone or after per
fusion with aldehyde fixative (Fig.2). During dehydration with alcohol or
acetone the tissue darkens, and the detailed pattern becomes indistinct. Therefore,
if trimming of blocks of osmium-treated tissue is wanted, it should take place
either in the osmic fixative or during the first step of dehydration. Obviously, well
preserved tissue exclusively should be used for analysis. Extremely gentle handling
of the tissue during dissection is essential to avoid mechanical damage. After
successful fixation by perfusion all tissue can be used. From pieces of brain fixed
by immersion in osmium tetroxide only tissue elose to the cut surface exposed to
the fixative is usable (A, Fig. 1). However, together with such material it is
sometimes advantageous to inelude some white unfixed tissue (C, D, E, Fig. 1).
This facilitates the distinction of well fixed from imperfectly fixed but yet
somewhat darkened tissue.
8 J. F. ALKSNE, T. W. BLACKSTAD, F. WALBERG and L. E. WHITE JR.:
The use of methacrylate limited the safe size of the blocks to fractions of a
millimeter unless special precautions were taken to avoid the development of
bubbles around the block. Embedding in epoxy- and polyester resins (Araldite,
Epon, Vestopal) is always successful
with blocks as large as a millimeter
or more in length. This facilitates
topographical orientation consider-
A
c
~.'
t$Y
500)1
Fig.1 Fig.2
Fig. 1. Sketches to show one procedure for selecting blocks containing tissue elements with a known orientation.
The rat brain is used as an example. A hemisphere has been divided with a horizontal cut into two pieces with a
razor blade and fixed in osmium tetroxide. One of these pieces it shown at A, after further subdivision into three
parts. Black dots indicate darkening due to the fixative. One of the three parts is a slice containing the area
desired for study. The pale, unfixed interior is indicated. A similar slice is shown in B. From the well-fixed edge
of the slice flat blocks (C, D, c, d) can be isolated, with a predetermined orientation of their upper edge in relation
to specific tissue elements. For instance, the upper edge of the block can be made parallel with or at right angle
to apical dendrites, fibre bundles, etc. In this drawing, pyramidal cells and wavy fibres crossing each other at
right angles are suggested. White, unfixed tissue can be included for orientation (C, D) or be trimmed off immedia
tely (c, d). The flat blocks are allowed to fall on their side on the bottom of the gelatin capsule (E) or other vial
used for embedding, thus defining the plane of the u\trathin sections in relation to tissue structures
Fig. 2. Diagram of the hippocampal region in the rat, based on horizontal silver-impregnated sections from amiddie
dorso-basal level. It reproduces the approximate appearance of the cut surface of the hemisphere after fixation
as generally done in the studies reported in this paper (cp. Fig. 1). The rectangular areas (A-D) roughly indicate
the size and location of blocks dissected out for electron microscopy. One granular cell and one pyramidal cell
have been drawn diagrammatically with dendrites, as demonstrable with the Golgi method. Regio superior and
regio inferior (separated by solid arrow) are subfields of the hippocampus proper. The asterisks indicate the
obliterated hippocampal fissure. Double arrows separate area dentata from hippocampus. The pear-shaped field
at C, delimited by a dotted line, in subiculum and regio superior is the approximate terminal area of a new pathway
mentioned in the text (p. 16). For list 01 abbreviations, see p. 9
ably since it means that more structures, i.e., more criteria for orientation be
come included in the polymerized block. Preparatory to the cutting of ultrathin
sections, thicker sections (1-3,u) can be made from the same block; here nuclei,
fibre bundles and other landmarks are recognized. Tissue which is irrelevant for
Electron microscopy ofaxon degeneration 9
the problem under investigation, and tissue which for technical reasons (size) can
not be included in the ultrathin sections, can then be trimmed off in a controlled
manner.
Ab breviations
alv the alveus (the hippocampal white matter)
as astroglial process
ax axon
ß nucleus ß of inferior olive
b normal bouton
cap capillary
Cl. claustrum
Coll. info colliculus inferior
Coll. sup. colliculus superior
D dorsal accessory olive
d dendritic branch
De dendritic shaft
d.1. dorsallamella of principal olive
d. m. c. c. dorsomedial cell column of inferior olive
d.c. dorsal cap of inferior olive
fb fibre bundle
gran the layer of granular neurons of the area dentata
L. left side
M medial accessory olive
m mitochondrion
mf the layer of mossy fibres (axons from the granular cells)
mol the molecular layer of area dentata
N.c. nucleus caudatus
N.r. nucleus ruber
or stratum oriens (layer of basal dendrites)
Pali. globus pallidus
presub the presubiculum (Roman numerals indicate its outermost layers)
Put. putamen
pyr pyramidallayer (somata of hippocampal pyramidal cells)
R. right side
rad stratum radiatum (apical dendritic shafts of pyramidal cells)
s synapse (axodendritic junction)
sp spine
S.r. substantia nigra
t neurotubule
v.1. ventrallamella of principal olive
'V. I. o. ventrolateral outgrowth
For other symbols, see text to separate figures.
Above, it has been assumed that areas of nervous tissue small enough to be
accommodated within one ultrathin section are to be studied. Probably the neuro
anatomist will, as often, encounter the opposite situation, viz., that an area or
nucleus is much wider than the diameter of an electron microscope grid. In order
to be able to carry out a systematic analysis of an area in this case, one can sample
a number of blocks from different levels within the area (Fig. 3) and study these in
succession.
To the neuroanatomist unfamiliar with these techniques it should be emphasi
zed that the selection of nervous tissue for electron microscopy is not a task
10 J. F. ALKsNE, T. W. BLACKSTAD, F. WALBERG and L. E. WHITE JR.:
significantly more complicated than the corresponding work for light microscopy.
It should not be conceived of as a serious obstacle to the adoption of electron
microscopy to a neuroanatomical problem when otherwise desirable.
The unstained thicker sections can be studied with a phase contrast microscope.
However, since the architectonic details of the tissue are more clearly revealed
after some kind of staining, this is recommended. Probably, most of the methods
described in the literature will serve the purpose. In this laboratory a drop of
staining solution (a strong solution of toluidin blue in 50% aqueous pyridin,
slightly acidified with acetic acid) is applied to the section.
V.1. o After 1-3 minutes a cover slide is pressed down on the section
o
o which now presents a distinct cyto-architectonic picture
o
combined with a weak myelin staining.
In order to recognize, on the electron microscope screen,
the topographical relations in the tissue, it is often helpful to
d.l. have, initially, very large sections. Later, the area of study can
be reduced as desired by further trimming of the block, e.g., if
serial sectioning of a specific area is wanted. To accommodate
very large sections and avoid disturbing bars, we have to a very
M large extent used specimen diaphragms ("grids") with one single
hole 800 fl in diameter!. The hole is covered with a film of form
Fig. 3. Diagram il- var reinforced with carbon. When lead stained sections are
lustrating the regions
of the ventrallamella used, no difficulty is experienced in obtaining sufficient contrast.
of the principaI olive
from which blocks In this laboratory the staining procedure of KARNOVSKY (1961)
are taken for further and particularly that of REYNOLDS (1963) have been used.
fixation in osmium
tetroxide. For ex- Mapping of terminal degeneration or of other features in large
planation, see text sections up to several hundred micra in width, regularly requires
much time, more often days or weeks than hours. For such
work it is absolutely necessary to have at ones disposal grid holders in which the
grid (or diaphragm) can remain for long periods, and which can be removed from
and reinserted into the microscope several times. Repeated removal of a diaphragm
with an 800 fl hole from a holder almost regularly leads to rupture of the film.
Although the storing of a lead stained section for weeks is not completely
harmless to the quality of the stain and of the subsequent electron micrographs,
the effects of storage probably are too small to be deleterious in such work as
mapping of terminal degeneration.
When large sections are studied in the electron microscope, orientation may
be facilitated by freehand sketches made during the work. However, a single,
high contrast electron micrograph of the entire grid or a major portion of it can
easily be obtained with some types of microscope, by setting the second condenser
lens at maximum, turning off the objective and using a wide bore projector, alone
or combined with the intermediate lens2• The objective aperture must be with
drawn. Such a micrograph also represents a useful map or guide during work
with serial sections. Cell nuclei, capillaries, and other conspicuous landmarks can
be discerned. On prints of such micrographs or on freehand sketches the distribu
tion of degenerated boutons can be entered. However, an easier and more accurate
1 Manufactured by LKB-Produkter AB, Stockholm, Sweden.
2 A Siemens Elmiskop Ib has been used in the present studies. Some of the micrographs
were made with a Tesla BS 242 electron microscope.