£3239 ”HMH””NIH!UHHUI“!INIHMHHHHWHWWW/l

ruesfii

 

This is to certify that the

thesis entitled

A STUDY OF THE ULTRASTRUCTURE OF THE DEVELOPING CUNEATE
NUCLEUS OF POUCH YOUNG OPOSSUMS
AFTER EARLY DEAFFERENTATION

presented by

E.—Michael Ostapoff

has been accepted towards fulfillment
of the requirements for

' ’7-
M" 5: degree in /'J/[//flé CG;

 

 

 

 

Major professor

Date /37/////’,/ ///i/

0-7639

 

 

OVERDUE FINES:
25¢ per day per item
RETURNING LIBRARY MATERIALS:

“5” 1 Place in book return to remove
"‘"” ‘ '- : charge from circulation records

 

é -“\\\‘
‘4 m“); '

 

 

 

 

A STUDY OF THE ULTRASTRUCTURE OF THE DEVELOPING CUNEATE NUCLEUS OF POUCH

YOUNG OPOSSUMS AFTER EARLY DEAFFERENTATION

By

Ernst-Michael Ostapoff

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

MASTER OF SCIENCE

Department of Psychology

1980

ABSTRACT
A STUDY OF THE ULTRASTRUCTURE OF THE DEVELOPING CUNEATE NUCLEUS OF POUCH
YOUNG OPOSSUMS AFTER EARLY DEAFFERENTATION
BY

E.-MICHAEL OSTAPOFF

Removal of the forepaw on the ninth day post-partum results in a forty
percent reduction in the volume of the cuneate nucleus neuropil of
opossums not seen following removal at twenty days post-partum. Electron
microscopic morphometric analysis comparing the deafferented with the
contralateral dorsal cuneate neuropils at 3, 6, 9, and 11 days survival
was conducted to determine the ultrastructural correlates of this volume
loss. No changes were found in the area of 13 profile categories nor in
the synaptic density or frequencies of synapse type observed. Evidence
supporting possible axonal sprouting on the deafferented side was found
on the basis of vesicle size and shape within pre-synaptic terminals.

It was concluded that the volume loss seen at the light microscopic level
due to the type of input modification used in this study results in a
neuropil development which is indistinguishable from the contralateral
side of the same animals during the time course included in this study

and within the parameters selected.

ACKNOWLEDGEMENTS

I give many thanks to Dr. John I. Johnson who suffered through the
earlier drafts of this thesis and who encouraged me to finally complete
it. I also thank the members of my committee, Drs. Charles D. Tweedle
and Glenn I. Hatton.

I also wish to acknowledge the assistance of Jerry Benjamin in
preparing the photographic materials contained within.

Thanks also go to Stephen J. Warach for the use of some of his
preliminary results and his help in caring for the opossums used in
this study.

Finally, I must thank my wife, Robin, whose understanding, not
to mention her generous financial support during the preparation of this

thesis made it possible.

ii

TABLE OF CONTENTS

LIST OF TABLES. . . . . . . . . . . . . . . . . . . . .

LIST OF FIGURES . . . . . . . . . . . . . . . . . . . .

INTRODUCTION. . . . . . . . . . . . . . . . . . .
MATERIALS AND METHODS . . . . . . . . . . . . . . . . .

Subjects . . . . . . . . . . . . . . . . . .
ANALYSIS I. PROFILE AREA MEASUREMENTS AND SYNAPTOLOGY.
METHODS . . . . . . . . . . . . . . . . .

Measurements . . . . . . . . . . . . . . . . . .

Synaptic Density and Classification. . . . . . .

Statistical Considerations . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . . . . . . .

Profile Classification Data.

Synapse Data . . . . . . . . . . . . .
DISCUSSION. . . . . . . . . . . . . . . . . . . .

Interpretaion Of Profile Categories. . . . . .

Area Measurements. . . . . . . . .

Synapse Data . . . . . . . . . . . . . . .

ANALYSIS II. SYNAPTIC VESICLE MORPHOLOGY. . . . . . . .

METHODS . . ... . . . . . . . . . . . . . . . . . . . .

RESULTS . . . . . . . . . . . . . . . . . . . . . .

DISCUSSION. . . . . . . . . . . . . . . . .

GENERAL DISCUSSION. . . . . . . . . . . . . . . . . .

iii

Page

vi

10
10
11
ll
13
13
14
l6
16
27
27
27
32
35
37
37
38
44

49

Page

LIST OF REFERENCES . . . . . . . . . . . . . . . . . . . . . . . 53

iv

Table

LIST OF TABLES
Page
Mean Profile Category Area . . . . . . . . . . . . . . . . 17

T-Tests Of Differences Between Contra (C) And Deaff
(D) Sides. . . . . . . . . . . . . . . . . . . . . . . . . l8

Synaptic Relationships . . . . . . . . . . . . . . . . . . 29

Frequencies Of Terminals Whose Mean W/L Ratio Is
Within One Block Of The Modal Block. . . . . . . . . . . . 45

Figure

LIST OF FIGURES
Page

Examples of most of the profile categories included in

this study are shown in the following four figures taken
from the sample micrographs, in a cropped and reduced
format. Figures 1 and 2 are taken from the 6 day survival
Contra side, Figures 3 and 4 from the 6 day Deaff side.
Beneath each micrograph is a line tracing of some of the
profiles in each micrograph. For clarity, the majority of
profiles in each figure have been omitted in these tracings.
During actual data collection, all the profiles in each
micrograph were traced, categorized, and the areas measured.
A general inspection of the figures will reveal little or
no qualitative differences in the neuropil components.

This was reflected in the quantitative measures, (see
Results, Analysis I). . . . . . . . . . . . . . . . . . . . . 19

Taken from Contra side, 6 day survival. Shown here are

the two categories of pre—synaptic profiles identified

on the Contra sides participating in synapses with differ-
ent types of membrane specializations. The synapse in the
upper center (closed arrow) is between a rather large At
profile and a large Mtr profile. The membrane specializa-
tion was judged to be symmetrical. The vesicle data means
:;SEM for the At profile in this case were WXL= 6.27 i .33
and W/L= .7355 i .0608, indicating a distribution of

rather small flat vesicles within this terminal. The
synapse in the lower right (open arrow) is between a Ved
and'Mtr profile with an asymmetrical membrane specialization.
This terminal had a distribution of flatter but somewhat
larger vesicles (W/L= .6894;: .0176 and WxL= 6.66 i .49)

than the previous At terminal. The two Mf profiles in the
center of the micrograph are probably glial in nature and
show qualitatively different packing densities in their
filaments. Bar equals 1 mm. . . . . . . . . . . . . . . . . 20

vi

Figure

Page

Taken from the 6 day survival Contra side. On the left
side (closed arrow) is an asymmetrical synapse between a
large At profile and smallthr profile. The vesicle
measurements showed this terminal to contain vesicles
slightly rounder, though still within the range of flat
vesicles, and larger than those terminals in Figure l,
(WXL= 7.07 i .31 and W/L= .7402 i .0270). Atypical
examples of Vel and.Vad profiles are also shown in the
upper left and center of the micrograph. The Vel profile
satisfies the criteria for the Mt category except for the
few vesicles and floccular background in the lower por-
tion of the profile. The Vad profile also satisfies the
criteria for the Mt category except for the polymorphous
vesicles in the lower tip. In both cases it was judged
that these were representative of transitional zones in
the processes where the ultrastructural characteristics
of the processes were changing. A smaller Vad profile

is shown in the lower center. Though smaller than most,
this profile demonstrates more typical Vad ultrastructure.
To the lower right is the profile in the Contra treatment
mentioned in the text as most closely conforming to
previously published reports of "late stage" electron
dense degeneration. Bar equals 1 mm. . . . . . . . . . . . 22

Taken from the 6 day survival, Deaff side. This figure
illustrates a possible transition profile between the

Vad and Vel categories (see Discussion, Analysis I). In

the center of the micrograph is a large profile with
characteristic large polymorphous vesicles. On this basis

it was placed in the Vad category. Areas in this profile

appear identical in ultrastructure to those of the pro-

file in the lower right, categorized as a Vel profile. . . . 24

Taken from the Deaff side, 6 day survival. In the lower
right is a symmetrical synapse between a Ved profile and
an Elf profile. This Ved terminal contained rather large,
round vesicles (wa= 8.02 i_.77 and W/L= .8091 i .0608).
In the upper left is a profile categorized as Mf because,
in spite of the absence of organelles, it was felt that
this profile conformed to the characteristics of astro-
cytic profiles previously published (see Methods, Analysis
I). Immediately adjacent to this profile is an area of
extracellular space, Xcs. One can note that this region
is bordered by only a single membrane layer belonging to
the surrounding profiles. The apparent contents of this
space was judged to be artefact. Bar equals 1 mm. . . . . . 26

\

vii

Figure

Page

Synaptic density expressed in number of synapses per

500 squared microns of tissue sampled on the ordinate.
Individual survival times are along the abscissa and to

the right are the treatment group means. Solid circles

are from the Contra group; open squares from the Deaff

group. . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

(A) Scatter plot of mean vesicle size versus shape

(Contra side). Shown is the distribution of axonal

terminal mean vesicle measurements for size (WXL) on

the ordinate versus shape (W/L) on the abscissa from

the Contra side samples. Each point represents the mean

value from 11 vesicles measured within each pre—synaptic
terminal computed as above. (B) Scatter plot of mean

vesicle size versus shape (Deaff side). Same as 6 A

but data is from the Deaff side samples. . . . . . . . . . . 40

(A) Frequency of mean vesicle data for shape (W/L) of
vesicles. Shown is the number of terminals along the
ordinate, whose mean vesicle data fall within each of 9
blocks along the abscissa, representing a range of the
values of W/L (shape estimate). Each block represents

a range of .025, with block 1 containing those terminals
whose mean vesicle data are less than .650, block 2 ter-
minals between .650 and .675, etc. Block 9 represents
those terminals whose mean vesicle data are greater than
.850. Solid bars are from the Contra side data, open
bars Deaff side data, all are treatment totalsa (B) Fre—
quency of mean vesicle data for size (wa) of vesicles.
Shown is the number of terminals, along the ordinate,
whose mean vesicle data for size (wa) fall between each
of 9 blocks along the abscissa, representing a range of
wa (size estimate). Each block represents 0.50 mm2 of
enlarged micrograph, with block 1 representing terminals
whose mean vesicle data are less than 5.5 mmz, block 2
terminals between 5.5 and 6.0 mmz, etc.. Block 9 repre-
sents terminals whose mean vesicle data are greater than
8.5 mm . Solid bars are total treatment frequencies of
terminals in the Contra side samples; Open bars those from
the Deaff samples. . . . . . . . . . . . . . . . . . .1. . . 42

viii

INTRODUCTION

This study addressed the question of how the removal of the peri-
pheral afferent organs affects the development of a target structure in
the central nervous system (CNS). Removal of the forepaw in pouch young
opossums before the fifteenth day post-partum (D15) results in an
approximately 40% reduction in the volume of the cuneate nucleus at the
light microscop (LM) level (Schreck and Johnson, 1978). This study
sought to investigate the ultrastructural correlates of this induced
volume loSs.

A recent study (Brenowitz, in press) has shown that at D9, the skin
receptors in the opossum forepaw are relatively immature and sparse in
number and although neurites were seen in hypodermal and dermal skin
layers, none were found in the dermal-epidermal interface region nor
were any observed to be in contact with any sensory receptors. Such
contacts were seen in D20 samples, by which age removal of the periphery
no longer affects the cuneate nucleus at the LM level (Schreck and John—
son, 1978). Removal of the forepaw at D9 must also disrupt the most
distal portions of the dorsal root ganglia (DRG) peripheral processes
but whether this damage results in loss of the DRG cells, especially
since it occurs prior to the establishment of synaptic contacts on the
receptors remains to be determined. However, removal of the receptor
sites and the distal segments of the DRG peripheral processes might
reasonably be expected to greatly alter the synaptic activity and perhaps
the growth of the DRG afferent fibers. This, in turn should result in

l

2
an altered development of the target structures, i.e. the cuneate nucle-
us neurons. This predicted altered development has been seen in the
cuneate nucleus at the LM level manifested by a 40% loss of nuclear vol-
ume. Neuronal cell counts of the cuneate nucleus, comparing deafferen-
ted with contralateral cuneate nuclei, have been inconclusive because
the cuneate nucleus has a poorly defined ventral boundary and chromato-
lytic (degenerating) neurons were not observed in large numbers at the
LM level (Schreck and Johnson, 1978). Thus it appears that although
nuclear volume decreased dramatically with early deafferentation, this
decrease may not be due to cell loss, but rather alteration in neuropil
development. It is this aspect in particular that the present study was
intended to investigate. If the cuneate volume loss following deaffer-
entation is not due to cell loss but to an alteration of neuropil devel—
opment, what are the ultrastructural correlates of this alteration?
After early deafferentation in this system, it has been shown physiolo-
gically that the remaining area of dorsal cuneate on the deafferented
side responded to the peripheral receptor fields left intact, i.e. wrist
and arm. That is, no silent, unresponsive regions of the cuneate nucle-
us on that side were seen following histological reconstruction of the;
electrode penetrations (Hamilton and Johnson, 1973).

As the present study was undertaken, it was expected that data
would result which might reflect the effects of one of several possible
hypotheses which could account for the volume loss seen in early deaf-
ferentation of the cuneate nucleus of the opossum but which was not pre-
sent after late deafferentation (D15+). The loss in area might result
from: 1) degeneration of DRG afferents; or 2) cell death via transneuéfl

ronal degeneration; or 3) cells which normally migrate to cuneate may

3
alter their eventual target site; or 4) atrophy and/or degeneration or
withdrawal of elements (dendrites) of the cuneate neurons post-synaptic
to DRG fibers; or 5) reduced or discontinued rate of proliferation,
differentiation, and/or growth of cuneate neurons; or to combinations
of the above.
1. Degeneration of afferents from the dorsal columns.

The cuneate neuropil volume loss resulting from deafferentation
prior to D10, not seen following D15 deafferentation could reflect a
critical period in the reaction of the DRG cells to destruction of their
peripheral processes. Were the reduced volume seen in the cuneate the
result of only DRG afferent loss, then DRG fibers and terminals must
occupy 40% of the neuropil in the normal cuneate. Preliminary results
indicated that processes which could be interpreted as axonal (of which
DRG fibers and terminals constitute at best, the majority) occupy only
20% or less of the neuropil at this stage of development. If deafferen-
tation results in only removal of DRG afferent profiles, this should
only result in a loss of at most an area equal to that proportion. In
any case, the design of this experiment was expected to be able to detect
any decrease in axonal processes. Were the loss of peripheral fibers a
major factor, one would expect to detect a substantial reduction in the
measured area of the axonal profiles, possibly followed, in time, by
an increase reflecting the development of inputs from other CNS areas.
Reactive changes, e.g. degeneration, would also be detectable in the
present study, but only if the appropriate survival times are included.
There are some indications in the literature that degeneration due to
deafferentation in developing systems occurs at a much faster rate than

in adults (e.g. Delong and Sidman, 1962).

4

2. Degeneration of cuneate cell bodies

There was a notable lack of chromatolytic cell bodies seen in the
LM study on deafferented cuneate development (Schreck and Johnson, 1978).
In fact, although there is circumstantial evidence that there is some
cell loss (e.g. there are greater numbers of cells at 16 days post-
partum than at adult age D120+, Ulinski, 1969), the uncertain nature of
the ventral boundary of the cuneate nucleus makes cell counts difficult
(c.f. Schreck and Johnson, 1978), and none of the LM studies on dorsal
column nuclei (DCN) to date have reported observing the kind of normal
cell death commonly seen, for example, in the ventral horn of the spinal
cord (e.g. Chu—Wang and Oppenheim, 1978) or the DRG (Hughes, 1973).
It may be that this nucleus does not undergo a large reduction in cell
numbers during development. If so, this might be significant in its re-
action to deafferentation. That is, if the normal maintenance of the
developing cuneate neuron does not depend on afferent innervation then
deafferentation may not result in the loss of cell bodies. Thus, not
only does there appear to be no transneuronal cell loss due to deaffer—
entation but also little if any normal cell loss, as normally seen in
many developing systems. Though not considered likely, if degeneration
of cuneate cell bodies contributes to the volume loss, its effect might
be detected by the present study. Cuneate neuron degeneration could
only be detected at the appropriate survival times, as cell counts are
unreliable in this system. However, through comparisons of the nuclear
and cytoplasmic areas of the deafferented and contralateralucuneate sam-
ples, it may be possible to infer hyper- or hypoplasia of cuneate cell
bodies on the experimental side and the time course of such changes

resulting from alteration of the primary afferent input.

5

3. Departure of cuneate cells (or failure to arrive): Abnormal Migration.

. Abnormal primary migration due to early removal of primary affer—
ents has been reported in the chick nucleus tangentialis (Levi-Mental-
cini, 1949, but recently not confirmed by Peusner and Mbrest, 1977), and
chick nucleus angularis (Parks, 1979), systems where cytoarchitectural
boundaries of the nuclei in question and neighboring nuclei are distinct,
making cell counts relatively reliable. Migration of cells in the early
development of the cuneate nucleus has been problematical. Autoradio-
graphic studies have indicated that the endothelial cells destined to
migrate to cuneate and gracile nuclei undergo mitosis over a period of
several days during the first week post-partum, (Johnson and Watson,
unpublished observations). Thus, migration may also take place gradu-
ally during the first week. 'Golgi preparations of normal cuneate
(Warach and Johnson, 1977) show the presence of appreciable numbers of
endothelial cells until the latter part of the first week (D5) and the
presence of migrating cells during the entire period of D3- 13. There
is a peak number of migrating cells in the period from D5 to D8, just
prior to the age of deafferentation in this study, so possible influ-
ences of deafferentation on migration must be considered. At the elec-
tron microscope (EM) level I have not been able to identify the ultra-
structural characteristics of the migrating cells in this system. I
have not found profiles with the characteristics of migrating cells in
the fetal monkey forebrain as reported by Rakic (1972). Given the
absence of some such identification criteria for migrating cells at the
EM level and the necessarily uncertain cell counts for the cuneate nucle-
us,%it is difficult to design an experiment to directly address the

question of abnormal migration. If abnormal migration is taking place

6

then it might result in seeing fewer nuclear and cell body profiles in
the samples from the deafferented cuneate nuclei when compared with the
contralateral nuclei. If deafferentation causes abnormal migration and
those cells which have already migrated into cuneate before deafferenta-
tion are able to develop reasonably normally, then despite the reduced
volume of the cuneate neuropil at the LM level after early deafferenta-
tion, it may be that the neuropil of cuneate on the deafferented side
will appear normal with respect to profile area measurements. That is,
if the volume loss seen at the LM level is a result of fewer cells
locating themselves in the dorsal cuneate but.those which do develop
in cuneate do so normally then one would expect no differences in the
areas of profiles in any representative cross section of the neuropil.
4. Dendritic.Atrophy

Atrophy and withdrawal of post-synaptic elements without loss of
post-synaptic neurons after partial deafferentation has been seen in
many systems, e.g. olfactory bulb (Mathews and Powell, 1962, Pinching
and Powell, 1971), lateral geniculate nucleus (Mathews et al., 1960).
chick brainstem auditory nucleus laminaris (Benes et al., 1977). In the
latter study, using relatively mature chicks (5-7 days post-hatch; approx—
imately 12-15 days after synaptogenesis) axotomy of the direct inputs to
nucleus laminaris neurons resulted in rapid (96 hours) and large (85%)
area loss to the affected dendritic tree due to atrophy without loss of
the nucleus laminaris neurons or the dendritic tree innervated by unaf-
fected afferents over the time course of this study. Assuming that
dendrites make up a major pr0portion of the dorsal cuneate neuropil,
(preliminary measurements indicated approximately 20%), atrophy of such

magnitude is clearly not evident when deafferentation takes place after

7

the critical period, post D15. This discrepancy may be due largely to
the fact that in transsection the afferents, virtually all the synaptic
endings contacting nucleus laminaris dendrites degenerated, while in the
opossum there exist other synaptic connections than those of the DRG
fibers which would survive DRG deafferentation, e.g. cortical, rubral,
and spinal (Martin and Dom, 1970, Martin et al., 1974). The presence
of other unaltered synaptic connections does not account for the large
nuclear volume loss if deafferentation occurs prior to D15. One possi-
ble explanation is that these other inputs to cuneate may develop in the
period between D9 and D15. If this is the case, then an analysis of
synaptic vesicle morphology might be expected to result in altered
numbers and/or proportions of terminals types, if these differ from the
DRG terminals, on the experimentally manipulated side when compared to
those on the contralateral cuneate side, as these newer terminals
establish contacts. Another possible hypothesis for the discrepancy is
that in this study, deafferentation occurs during the period of initial
synaptogenesis and before full neuropil development, while in the
previously mentioned studies, the subjects underwent experimental mani-
pulation at more mature stages of development. It was anticipated that
atrophy and/or withdrawal of cuneate neuron processes would be manifes-
ted by either profiles exhibiting ultrastructural characteristics of
atrophy or degeneration, or in reduced measured area of dendrite-like
processes in the samples on the deafferented sides when compared to the
control sides.
5. Altered proliferation, differentiation and growth

The differential effects of deafferentation during the critical

period versus after the critical period may be an interruption of and

8
suspension of the normal developmental ontogeny of the cuneate neuropil.
Golgi preparations have been utilized as one method of examining the
response of developing neurons to deafferentation, however the difficul-
ty of obtaining litters.of pouch young opossums of known ages combined
with the erratic nature of Golgi impregnation argued against using those
methods. In addition, any reactive alterations in the cytoplasmic
organelles or membrane specializations of cuneate cells, and alterations
in the incoming DRG fibers and terminals which possibly might occur due
to deafferentation could not be visualized using standard Golgi impreg-
nated material. Gold-toning impregnated and identified cuneate neurons
for EM analysis might be a viable, informative methodology, but repeated
attempts in this laboratory over the past three years have not yielded
a single impregnated cell in glutaraldehyde fixed pouch young opossum
material (in contrast to 5—10 day old chick and adult raccoon material).
The identification of all the individual profiles of a cross section of
the neuropil was expected to result in a composite picture of the rela-
tive complexity of that neuropil and provide a basis for comparisons
of the relative complexity between deafferented and normal cuneate neuro-
pils much as one would compare branching patterns and dendritic lengths
in Golgi preparations but without the vagaries of the latter method.
During the normal development, the area of dendrite-like profiles increa-
ses with age (Tweedle et al., 1977), i.e. as the dendrites of the cuneate
neurons grow and proliferate, they occupy a larger proportion of the
cuneate cross sectional area. If the normal differentiation of the
cuneate neurons is interrupted as a result of early deafferentation, then
it was expected that there would be an alteration in this increase in

area. In addition, there was expected to be a qualitative difference

9
between the appearance of the neuropil on the deafferented and the
control sides. During the normal development of neurons, the few, large,
relatively sparsely branched processes of the immature neuron give way
to smaller, more profusely branched processes of the adult neuron. If
deafferentation results in arrested differentiation of the cuneate neu—
ron, then a representative cross section from the neuropil of the
deafferented cuneate should contain relatively few small processes and
relatively more large ones. Thus, this study was expected to be able to
indicate the time course and relative complexity of development of the
cuneate neurOpil on both the deafferented and contralateral sides of the
brainstem.

An alteration in trophic factors might account for this possible
arrested development. It might be that although removal of the forepaw
does result in altered, presumably reduced, synaptic acitvity of the DRG
terminals, damage to the DRG peripheral processes may be insufficient to
cause either cessation of some undetermined neuronal maintenance factor
not directly linked with synaptic transmission or atrophy related to
terminal and fiber degeneration and concomitant glial engulfment of
degenerating processes. If this is the case, then analysis of synaptic
terminal vesicle morphology should indicate the continued presence of DRG
fiber terminals, again only if they constitute a recognizable sub-
category. If the DRG fiber terminals remain and the normal neuropil
morphology does not develop, then it would seem evident that it is the
influence of active peripheral afferents, either their active presence
at a developmentally crucial period or their release of some trophic
factor dependant on the DRG fiber activity which is the active agent in

cuneate neuron development. The role of the remaining unaffected

10
synaptic inputs (cortical, etc.) may then be considered as maintaining
a level of development rather than stimulating new or furthering devel-
opment.

The present study will be reported in two parts. Analysis I will
include the data from the measurement of profile category areas and the
density, type, and distribution of synapses. Analysis II will include
the vesicle morphology data from the pre-synaptic axonal terminals

identified in Analysis 1.
MATERIALS AND METHODS

Subjects

Pouch young opossums, Didelphis virginiana, of known ages were sub-
jects of this experiment. The ages of subjects were determined by daily
checking the pouches of female opossums bred in captivity. D1 was taken
as the first day in which pouch young were observed. 0n D9 all litter-
mates had one of their forepaws amputated without removing them from the
mother's nipples while the mother was lightly anesthetized with sodium
pentobarbital (20 mg/kg). Survival times of 3, 6, 9, and 11 days were
sampled. On the appropriate day of sacrifice, one subject was removed
from the mother's pouch, weighed, measured, and anesthetized with an in-
traperitoneal injection of 0.005 mg sodium pentobarbital and 0.095 ml
.9% saline. They were then sacrificed by intracardial pressure perfusion
with 2% paraformaldehyde, 2% purified glutaraldehyde in .1M sodium caco-
dylate containing 0.5% DMSO and 0.001% CaCl2 (pH 7.5 and 600 mOS). The
skull was then removed over the midbrain and obex and the whole animal

immersed in the fixative overnight at 40 C. The medulla was then blocked

11
at the obex into a 1 mm post-obex slice which was divided into right and
left halves, rinsed in .1M cacodylate and postfixed overnight at 40 C or
8—10 hours at room temperature in 2% 0s04- 2% potassium ferrocyanide
(Dvorak et al., 1972). The blocks were then rinsed in .lN sodium ace—
tate and stained in block with 1% uranyl acetate for at least 2 hours at
room temperature or 8 hours at 40 C and then dehydrated through a graded
series of alcohols and embedded in plastic for electron microscopy.
Semi-thick (1-2 m) were cut from both halves of the brainstem and
stained with 2% methylene blue to localize the cuneate nucleus. Ultra-
thin (60 nm) sections were then cut and mounted on uncoated 200 mesh
copper grids. Using the semi-thick sections for reference, the dorsal
aspect of the cuneate nucleus was located on the grids under the electron
microscope (Philips 201) and a series of 16 nearly contiguous but
non-overlapping negatives were taken at a primary magnification of
15,000. These were then enlarged 2.8 times and printed, yielding a

final magnification of 42,000.
ANALYSIS I: PROFILE CATEGORY AREA MEASUREMENTS AND SYNAPTOLOGY
METHODS

These enlarged micrographs were then analyzed in the following
manner. Individual profiles were assigned to categories which were
established by noting the organelle contents and ultrastructural charac-
teristics of the various profiles in the samples. If further analysis
showed a reasonable frequency of occurrence of any particular profile

type, that category was retained. The categories included:

Nuc-

Cyt-

Mtr-

Elf-

Mt-

At-

Vel-

Vad-

12

profile containing chromatin, defined as a "fairly homogenous
population of fine strands" (Peters et al., 1976 pp. 47-50).
May contain one or more nucleoli (dense, large, roughly
spherical mass clearly demarcated from the rest of the karyo-
plasm). The profiles included in this category are generally
regarded as representing the nuclei of neurons (Peters et al.,
1976 pp. 47).

profile adjacent to Nuc; irregular in shape; containing poly-
ribosomes, mitochondria, some microtubules, lysosomes, Golgi
bodies, rough and/or smooth endoplasmic reticulum (ER). This
category is generally understood to represent the cell bodies
of neurons (Peters et al., 1976 pp. 11).

profile with a regular shape containg microtubules and ribo-
somes; can contain smooth ER. This category is generally
understood to represent the dendritic processes of neurons
(Peters e. al., 1976 pp. 10).

profile with an electron lucent background and a floccular
contents; can contain a few pleomorphic vesicles and have
microtubules at one end; profile is usually large with a
regular shape and occasionally will have a club-shaped expan—
sion at the other end. This category closely fits the des-
cription of dendritic growth cones (Hinds and Hinds 1976).

profile with a regular shape containing microtubules but no
ribosomes; can contain smooth ER. This category is generally
taken to represent unmyelinated axonal cylinders and/or the
distal most tips of dendrites (Peters et al., 1976 pp 10).

profile with an electron lucent background filled with synap-
tic—sized vesicles (40-50 nm in diameter) and mitochondria.
This category corresponds to the classical axonal terminal
(Peters et al., 1976 pp. 10).

profiles, usually large, containing floccular material and
synaptic-sized vesicles, usually few in number and/or clumped
together. Hanaway and Smith (1979) described profiles similar
to these as "light type"degeneration. Gilbert and Stelzner
(1979) found profiles like these in normal material from
developing rat spinal cord. Falls and Gobel (1979) described
profiles such as these as both degenerating and extending.

large, electron lucent processes containing mainly smooth ER
and clear vesicles, larger than 40—50 nm with thinner mem-
branes than synaptic vesicles. The profiles in this category
are similar to those Falls and Gobel (1979) described as con-
taining addition vesicles and were in the process of either
extending or withdrawing.

l3

Ved- electron dense processes with clear or electron dense vesicles
40-50 nm in diameter. May contain mitochondria and/or micro-
tubules. This category as well as Ed might be taken as simi-
lar to what has been described as "dark type" degeneration
(e.g. Gilbert and Stelzner 1979, Okado et al., 1979).

Ed- electron dense, small processes (many are finger-like in shape)
with no discernible organelles or a diffuse mixture of organ-
elles.

ME- any profile containing microfilaments or any small, finger-

like, electron lucent process which appears to fit the spaces
between other processes. This category has some of the fea-
tures others have ascribed to glia, usually astrocytes (c.f.
Peters et al., 1976 pp. 233-244).

Unid- unidentified profiles, i.e. those lacking criterial organelles
or shape either because they were too small or on the edge
of the micrograph.

Xcs- extracellular space; an area without organelles bounded by
the membranes of surrounding profiles.

Measurements

After classification, the area of the profiles was measured using
either a 1 cm or a 0.5 cm grid overlay (Weible and Bolander, 1974), de—
pending on the size and shape of the profile. The areas of large, regu-
larly shaped profiles could be accurately approximated using the larger
grid. It was felt that smaller profiles, or those with irregular shapes,
would be more accurately measured using the smaller grid. Areas were
then totalled within each category for each sample of 16 micrographs.
The actual total area of tissue examined in each sample was computed in
the following manner. In a micrograph, 1 cm2 equals (42,000)2 / 10
umzl cmz, therefore since each micrograph equalled 1824 cm2, a sample
of 16 micrographs equals 29184 cmz. This represents approximately 1654
“m2 of tissue.
Synaptic density and classification

Synaptic density (synapses per 500 um2 area sampled) was measured.

14

A synapse was defined as a paired membrane specialization (thickening)
of opposed membranes with one of the included profiles, defined as pre-
synaptic, containing electron lucent vesicles 40—50 nm in diameter in
close appostion to the membrane specialization. The category of pre-
and'post-synaptic (the latter defined as the other profile opposite
from the pre-synaptic profile) profiles was also tabulated. In addition,
the membrane specialization was classified as asymmetrical (AS) if the
post-synaptic specialization was thicker than the pre-synaptic one or
symmetrical (SS) if the pre-synaptic specialization was thicker or if
the two appeared approximately equal in thickness.
Statistical considerations

Because only one animal per age was sacrificed, the only valid
statistical comparisons which can be made with any of these data is the
one considering the effects of treatment, i.e. contralateral (Contra)
versus deafferented (Deaff) sides. A t-test for related samples was
chosen as the best test since the Deaff and Contra samples were obtained
from the same animals. No overall statistic was used because the compar-
isons of interest are only those between specific categories within a
single animal, Contra versus Deaff sides. Some of the profile categories
included in this study have ultrastructural characteristics similar to
those that previously published reports suggested were indicative of the
onset of some important underlying process. Before analysis of the data,
some of the categories were chosen for statistical comparison on the
basis of these published reports. This was done to avoid making all
possible comparisons and "hunting" for a significant difference with the
resultant loss of statistical power from conducting too many tests. The

categories chosen were:

15

Elf (possible growth cones). If deafferentation affects growth
rates, than the Deaff side could be expected to have fewer of these
profiles.

Vel (possible "light type" degenerating axonal terminals and/or
growth cones). This profile category, if symptomatic of degeneration,
would be expected to increase in area on the Deaff side, relative to the
Contra. If this category represents growth, then the Contra side might
be expected to show a larger area of this type of profile.

Vad (possible "addition vesicle" containing profile). The same
rationale as for the Vel category was used in including this category
for statistical testing.

At (axonal terminals). Deafferentation might be expected to result
in fewer of these profiles on the Deaff side relative to the Contra side.

Ved, Ed (possible "dark type" degeneration). If these categories
represent degeneration, then the Deaff side would show an increase in
area of thesetypes of profiles over the Contra side.

Mf (astrocytic elements). Gilbert and Stelzner (1979) describe a
reactive astrocytic engulfment of degenerating neuronal elements in
neonatal material. If this occurs in the present case then the Mf
category would be expected to have a larger area on the Deaff side than
on the Contra side.

Mtr (dendrites). Early deafferentation has been shown to reduce
the size, number of spines and branching of the dendrites of developing
neurons (see Cowan, 1970, review). If this occurs in the present study
then areas of the Mtr category on the Deaff side would be expected to
be reduced from that on the Contra side.

Xcs (extracellular space). Falls and Gobel (1979) described

16
"cavitation of degenerating processes" rather than glial engulfment in
material from neonatal kittens. If this process were evident in the
present study, then one might expect the Xcs on the Deaff side to

increase in area relative to that on the Contra side.
RESULTS

Profile classification Data

The mean area for each category by survival time and treatment
(Contra versus Deaff) is shown in Table 1. Brief examination of the
means within each column indicates that no consistent differences
between the treatments were found. This observation is borne out by the
t-tests for related samples performed on the data for those categories
mentioned above in which one might, on the basis of previously
published reports, expect to see differences in area resulting from
removal of the periphery. The calculated t's are given in Table 2.
Since t = 3.182, one can readily see that none of the comparisons

.05; 3df

approaches significance. Note that since the values for the Deaff side
were subtracted from those on the Contra Side, a positive value for the
t statistic indicates that the area of a particular category from the~
Contra side was greater than that from the Deaff side and vice versa.

Figures 1 through 4 demonstrate that in addition to no quantitative
differences, the Deaff and Contra cuneate samples showed a remarkable
lack of qualitative differences as well. Figures 1 and 2 are taken from

the 6 day survival Contra side and Figures 3 and 4 from the 6 day survi—

val Deaff side.

17

we see mm mm wmm emm mm: om: me mo mme ewe cam:

moi on me mm mmm mam mmm mmm mm . on mam mop La
mm mam so .mw so: mam mam mm: on 0» mm was m
mm om see so 1mm awn as: mam mm mm we a: m
we me off msa cam mom cm: was mma m» sea amp m
a o a u a o a o a o a o
wax was: use 0:2 m: we
amp one man wmm a» me am mo as mm em mm we so use:
nae smi moi nma emf am is . am am mm mm em am use me
a .2 on. Rm 5 R i a S we Om mm mm B 3 a
emf mma mam mom m: mm am so moi mm o: in mm me o
one moi ewe mam we em we _ a: mm om mm of Hoe A» m
a u a o a u a o a o o u a o
L: up: em> u< em> Hm> «Hm

muowoumo mfifiwoum

.Aosmmqu mo E: poem:Vm u Pom. moEHu mfinmu Eouw Hwesscv

Hemoww cofimuo>coo w wcfim: as osmmfiu mo E: ponmsvm om wouuo>coo we use ofinmu osu :« necessz .mmpwm «upon
mam Eouu : ”mopfim whucou wen Eouw cum 0 maisfioo c“ mama .asouw ucmsumouu some you consumes zuowoumo
mfiwuoue mo AEo poemsvmv moan sees as» geomouawu :cmoZ: m3on may a“ necessz .ofimsmm some aw popswmwe
seeHwODUfiE no so pwumscm mo Hosea: sews or» ucowoueou magma mus» a“ HH .9 .m .m mzou ecu cw muwn532

m<mm< VMODMH<U demomm z<mz .# m4m<fi

TABLE 2.

l8

T-TESTS OF DIFFERENCES BETWEEN CONTRA (C) AND DEAFF (D) SIDES

Shown are t-tests for related samples of the selected categories (see

Given here are the categories, means
of the differences between each related pair, the standard deviation of
those differences, and the t values (Roscoe, 1969).

Methods for selection rationale).

Category

Elf
Vel
Vad
At
Ved
Ed
Mf
Mtr

Xcs

Mean of
Differences
in Area (cm )
(C-D)

17.69

-O.62

-20.15

11.78

-2.52

0.93

-5.22

1.47

21.89

Standard Deviation
of the Differences

35.79
11.23
26.28
13.35
11.41
61.14
17.80
41.62

52.32

.494

.055

.767

.882

.221

.015

.293

.035

.411

19

Figures 1 through 4. Examples of most of the profile categories

included in this study are shown in the following four figures taken
from the sample micrographs, in a cropped and reduced format. Figures
1 and 2 are taken from the 6 day survival Contra side, Figures 3 and 4
from the 6 day Deaff side. Beneath each micrograph is a line tracing
of some of the profiles in each micrograph. For clarity, the majority
of profiles in each figure have been omitted in these tracings. During
actual data collection, all the profiles in each micrograph were traced,
categorized, and the areas measured. A general inspection of the figures
will reveal little or no qualitative differences in the neuropil compo-
nents. This was reflected in the quantitative measures (see Results,
Analysis I).

Figure 1. Taken from Contra side, 6 day survival. Shown here are the
two categories of pre-synaptic profiles identified on the Contra sides
participating in synapses with different types of membrane specializa-
tions. The synapse in the upper center (closed arrow) is between a
rather large At profile and a large Mtr profile. The membrane speciali-
zation was judged to be symmetrical. The vesicle data means : SEM for
the At profile in this case were WxL= 6.27 :_.33 mm2 and W/L= .7355 i
.0608, indicating a distribution of rather small flat vesicles within
this terminal. The synapse in the lower right (open arrow) is between

a Ved and Mtr profile with an asymmetrical membrane specialization.

This terminal had a distribution of flatter but somewhat larger vesicles
(W/L= .6894 i .0176 and WxL= 6.66 :_.49) than that of the previous At
terminal. The two Mf profiles in the center of the micrograph are
probably glial in nature and show qualitatively different packing den-
sities in their filaments. Bar equals 1 mm.

 

 

 

Elf

Elf

 

Figure 1.

 

21

Figure 2.

Taken from the 6 day survival Contra side. On the left side (closed
arrow) is an assymmetrical synapse between a large At profile and small
Mtr profile. The vesicle measurements showed this terminal to contain
vesicles slightly rounder, though still within the range of flat vesi-
cles, and larger than those terminals in Figure l, (wa= 7.07 i_.3l and
W/L= .7402 i .0270). Atypical examples of Vel and Vad profiles are also
shown in the upper left and center of the micrograph. The Vel profile
satisfies the criteria for the Mt category except for the few vesicles
and floccular background in the lower portion of the profile. The Vad
profile also satisfies the criteria for the Mt category except for the
polymorphous vesicles in the lower tip. In both cases it was judged
that these were representative of transitional zones in the processes
where the ultrastructural characteristics of the processes were changing.
A smaller vad profile is shown in the lower center. Though smaller than
most, this profile demonstrates more typical Vad ultrastructure. To the
lower right is the profile in the Contra treatment group mentioned in
the text as most closely conforming to previously published reports of
"late stage" electron dense degeneration. Bar equals 1 mm.

     
 
 

  

‘. ’."
. ‘r.‘\ u_:..
‘3‘ ,
u; a‘ #{j
' SLI'

Eire” 1' a!

  

Lu. ,
1"" '

‘1, 1'. _ "2,;. L.
.3 5:31;:{fl “Sfifitta‘ém

 

Vel

Vad

Mfr Ed

 

 

 

Figure 2.

 

23

Figure 3.

Taken from the 6 day survival, Deaff side. This figure illustrates a
possible transition profile between the Vad and Vel categories (see
Discussion, Analysis I). In the center of the micrograph is a large
profile with characteristic large polymorphous vesicles. On this basis
it was placed in the Vad category. Areas in this profile appear identi-
cal in ultrastructure to those of the profile in the lower right, cate-
gorized as a Vel profile. Bar equals 1 mm.

24

. .l.: l! fig,‘ .1
. ...m .2 4

..,...K .L .1“ 3h
. ..q. ifivm. r w

   

.4,

         

w. 3"
'1'
r4

I'DV#”I&
3 “Gig? 3

  

  

... .... $43...
(r.;.~ in
w

1,. .. 4
4353...: .
.. . .

_,. a. ... . m. A

1' A

 

T?

    

e

   

        

, _

     

. .u. a.
. .,......n.w . . .
.. .. .. we. at“?_.,,..........,.,.,v..... ....

 

 

 

Vel

UC

 

 

Figure 3.

25

Figure 4.

Taken from the Deaff side, 6 day survival. In the lower right is a
symmetrical synapse between a Ved profile and an Elf profile. This Ved
terminal contained rather large, round vesicles (WxL= 8.02 i .77 and
W/L= .8091 i .0608). In the upper left is a profile categorized as Mf
because, in spite of the absence of organelles, it was felt that this
profile conformed to the characteristics of astrocytic profiles previous—
ly published (see Methods, Analysis I). Immediately adjacent to this
profile is an area of extracellular space, Xcs. One can note that this
region is bordered only by a single membrane layer belonging to the
surrounding profiles. The apparent contents of this space was judged
to be artefact. Bar equals 1 mm.

26

 

 

 

Mf

Xcs

Cyt

Ved
El

Nuc

Ed

Cy;

 

 

Figure 4.

27
Synapse Data

As anticipated, there were too few synapses in the samples (6-10
each) to justify considering these data statistically. The samples from
the Contra side had a total of 32 synapses for a mean of 8 per sample.
In the samples from the Deaff sides, 31 synapses (mean 7.75) were iden-
tified. Figure 5 shows the synaptic density for the two treatments, both
by sample and by total for all samples in a treatment. No obvious
differences exist.

Table 3 shows the frequency of each type of pre- and post-synaptic
element and their relationships to each other; i.e. the pre- and post-
synaptic element pairings.. No outstanding differences between the Contra
and Deaff samples was noted, e.g. the greatest percent difference in
Table 3 is that between Ved-E1 type synapses. A t-test for related
samples was performed and found non-significant (t= —0.44, Contra versus
Deaff). For convenience, and because of the relative paucity of Vel and
Vad profiles participating in synapses as the post-synaptic element, the
category "electron lucent" (EL) was included in the post-synaptic ele-
ment lists. This was defined as Elf plus Vad (2 profiles in the Deaff
samples) plus Vel (1 profile in the Contra samples). Three Vel pre-
synaptic elements were identified from the Deaff sides, and were inclu-
ded in Table 3. None were identified in the samples from the Contra

sides.

DISCUSSION

Interpretation of profile categories

Most of the profile categories based on organelle contents used

28

4.0?

9‘

C)
1
o

Synopses Per 500nm2
_ N
b o
l 1
o o\
/G
/

 

 

, 1 1%.

6 ' 9 II Group
Totals

(”-1

Survival Time (Days)

Figure 5. Synaptic density expressed in number of synapses per 500
squared microns of tissue sampled on the ordinate. Individual sur—
vival times are along the abscissa and to the right are the treat-
ment group means. Solid circles are from the Contra group; open
squares from the Deaff group.

29

SYNAPTIC RELATIONSHIPS

TABLE 5.

DEAFF

CONTRA

$ owuamczm
umoa Hence

Mama umomuoua $
is
owuamcxm umoe

max“
ofiuamcmm mum

fi owummaxm
mum Hmuou

& owuamazm
umoa Hmuou

Mama uncanoum &

a

ofiuemczm umoa

.oamu
ofluamcam one

fi ofiuamcxm
mum Hmu0u

10

E1

9
5

El

45

10

11 51

0

E1

El

9
.5
6
1|.
7) 22
l
) \I/
r r
( (
t t
M M
. _
S S
a S
. .
t
A
2
2/
7|
.4
l
2
2/ O/
l
) )
r r
( (
t t
M” M
_ .
S S
a S
. .
t
A
‘4
14

5

Ed

5

Ed

10

16

0

Ed

0

Ed

Cyt 5

Cyt 5

Cyt O

Cyt 5

E1 6
29

“3.5“

19

15
6

El

25

E1

E1

Mt(r) 15

Mt(r) 9

25

Mt(r) 10

Mt(r) 16

58 Ved

S6 Ved

6

Ed

6
6

Ed

12

Ed

Ed

Cyt 0

Cyt O

Cyt 0

Cyt 0

5

E1

5

El

10 Vel -ss-

5

Cyt 5

-SS-

30

in this study have definite and clear interpretations in the literature.
For instance, it is fairly obvious that the categories Nuc and Cyt
represent the nucleus and cytoplasm of what are generally taken as neu—
rons. Similarly, the categories Mtr, Mt, and At represent dendrites,
(mainly proximal ones), axonal cylinders and/or smaller dendritic ele-
ments, and axonal terminals, respectively. For these categories the
best reference is Peters et al.(l976).

For the three electron lucent categories (Elf, Vad, Vel) estab-
lished in this study an interpretation based upon previous literature
is not so clear. Many authors have described profiles with characteris—
tics similar to those of the Elf category as dendritic growth cones (e.g.
Hinds and Hinds, 1976, Skoff and Hamburger, 1974, Juraska and Fifkova,
1979, and Okado et al., 1979). However, several authors have described
similar profiles and termed them "large axonal endings" (Singer et al.,
1979), "dendritic spines" (Privat et al., 1974), and "Bergmann glial
processes" (Del Cerro and Snider, 1972).
The three pre-synaptic Vel profiles seen in the Deaff samples lend
support to the supposition that the vesicles within these profiles are
indeed synaptic vesicles. The Vel category may represent the growing
tips of axons if the floccular matrix is an indication of an active
growth process. However, large electron lucent terminals, with either
a floccular or filamentous background matrix and clumped vesicles have
been interpreted by several authors as early degenerative stages (Cook
et al., 1974, Mathews et al., 1977, Hanaway and Smith, 1979, Ralston and
Ralston, 1979) but this has been disputed by Gilbert and Stelzner (1979)
who found these profiles in reasonable quantities in normal material and

they concluded that they were probably not degenerative.

31

The Vad category has many of the ultrastructural characteristics
of the profiles described as either growing and/or degenerating by Falls
and Gobel (1979) in the substantia gelatinosa of the kitten. These
authors suggest that the ultrastructural manifestations of growth and
degeneration or retraction of processes during development may be
identical; or at least some of the stages of these two processes may
appear identical. This leads to the supposition that the mechanics of
growth and degeneration may be similar processes, at least in neonates.
For instance, degeneration may proceed using some of the same stages
but in the reverse order of developing or elOngating processes. If this
is the case, then all three of these electron lucent categories may be
either developing or degenerating profiles, depending on the ultimate
disposition of the parent cell. It should be noted that there is consi-
derable possibility of overlap between these three categories as the
distinguishing features separating them may or may not be present in
every cross section through a particular process. In addition it is
possible that these categories may represent subsequent or previous
stages of one another, depending upon whether that specific process is
elongatingcn: retracting. Without detailed serial section analysis,
clear functional interpretation of these categories is really not
possible.

The electron dense categories (Ved and Ed) also present interpre—
tational dilemmas. Many authors have described similar profiles as
"dark type" degenerating processes (e.g. Gilbert and Stelzner, 1979,
Okado et al., 1979, Oppenheim et al., 1978). Electron density has also
been used to characterize, in conjunction with other criteria: axonal

growth cones (Skoff and Hamburger, 1974); dendritic growth cones (Skoff

32
and Hamburger, 1974, Privat et al., 1976); macrophages (Lenn, 1978a, b).

and oligodendroglia (West and Del Cerro, 1976). With reference to the
glial elements, no cell bodies characteristic of oligodendroglia or
microphages were encountered in the sample micrographs. This is not to
say that some of these electron dense processes could not be glial in
origin. As to the dark growth cones versus degeneration, there was
very little seen in the sample micrographs from which to form a conclu-
sion. Only one "late stage" degenerative terminal (defined as very
electron dense with swollen vesicles, see Gilbert and Stelzner, 1979)
was seen in each of the treatment groups. One of these, that from the
Contra treatment group is shown in Figure 2.

The Mf category included all those profiles which contained micro-
filaments plus the small sinuous profiles most authors agree are proba-
bly glial (primarily astrocytic) in origin (c.f. Peters et al., 1976
pp. 233-244).

Area Measurements

The absence of any differences in the areas of the various profile
categories was disappointing. Ther are several hypotheses which might
account for this finding.

. One explanation might be that due to the small number of samples,
the pooling of data across all survival times may have effectively hid-
den any effects. That is, any transient differences, i.e. for instance
at the earliest survival time in this study (3 days), might be submerged
by combining those data with the data from the other survival times.
For instance, if one subtracts the Deaff side from the Contra side at
each Survival time, the Elf and Vel categories all show a nominal trend

from a large negative value (less area in the Deaff samples) at three

33

days through smaller negative values at each of 6 and 9 days to an abrup-
1y large postive value at 11 days survival. The At category, also, is
initially a large negative value at 3 days survival. But this category
then, at 6, 9, and 11 days shows an increasing postive difference between
the Contra and Deaff sides. A few of the other categories show similar
trends. Of course, in looking at the data in this fashion, it is inap-
propriate to apply any statistical tests, both from the standpoint of
too few degrees of freedom as well as the post hoc nature of this inquiry.

Another possible hypothesis for the lack of differences seen in the
measurements of area may be that the survival times chosen in this study
were inappropriate. There is some evidence, albeit in different systems
using more direct lesioning methods, that some changes due to alterations
of input to developing systems occurs over time periods of 1-2 days
(Delong and Sidman, 1962). Other changes (e.g. sprouting) in mature and
developing systems occur over time periods equal to or in excess of the
longest survival time used in the present study, 11 days. One might
present the case that the majority of degeneration or withdrawal effects
of early deafferentation in this system had already occurred prior to
the initial survival time used in the present study. The nominal dif-
ferences mentioned above could be taken as supporting this hypothesis.
These differences between the Contra and Deaff sides, at the early sur-
vival times, could represent residual effects lingering on after the
majority of the modifications due to deafferentation had already taken
place and dissappeared.

Two other hypotheses seem possible to explain the profile classifi—
cation data. It appears possible that abnormal migration (c.f. Parks,

1979) might occur in reaction to early deafferentation resulting in

34
fewer cells being located on the dorsal aspect of the cuneate nucleus.
This hypothesis would have to assume that the cells remaining in the
cuneate nucleus would develop similarly to the Contra side with regard
to the kinds of measurements used in this study. This latter assumption
would also be inherent in an alternative hypothesis to explain these
data; namely that even with equal numbers of cells on the Deaff and‘
Contra sides, the Deaff side could hypothetically develop fewer neuroPil
components (dendrites, axpns, etc.) although all these components could
be in similar pr0portional areas to those of the Contra sides. Since the
samples taken from each of the sides were equal in area, this would
result in a larger percent of the entire neuropil of the Deaff side being
included in the samples but would not result in any changes in the areas
of the various profile categories. For illustration; if the neuropil on
the Deaff side had a total area which was 60% of the Contra side but
each of the categories of profiles in that Deaff neuropil were represen-
ted in equal proportions with the Contra side (i.e. 60% of each profile
category in the Contra side) then samples taken from each of the sides
which were not totally inclusive of all the neuropil would not reveal
any differences in the area of profile classifications. If the Deaff
side were 60ljm2 in cross sectional area and the Contra side were 10011m2,
then samples of 40 um2 taken from both sides would include 40% of the
Contra neuropil but 67% of the Deaff side neuropil. In the absence of
any other differences in the neuropil elements other than proportional
numbers of elements, no differences in the areas of the categories would
be expected. I would like to feel that these latter two assumptions of
proportional develpment even without normal input is less likely than

the effects of deafferentation occurring either earlier than the 3 day

35

survival period used in this study, or the effects may have been hidden
in the treatment pooling. If the assumption is true that the volume

loss seen at the LM level due to the type of afferent input modification
used in this study results in a neuropil development which is proportion-
ately smaller but otherwise normal (at least within the parameters used
in this study) at the sub-microscopic level, then it is an example of
interesting and potentially important data which is statistically non-
significant and thus not publishable in this form.

Synapse data

One explanation, alluded to in the Results section above, for the
lack of differences seen between the Contra and Deaff sides in these
data is that this could be an anomaly resulting from too small a data
base. The only remedy for this would be to increase the sample size to
ensure that a sufficiently large number of synapses were present. This
involves what I consider to be a prohibitively large expense in material
and effort.

An alternative is that there is no effect on the synapses in the
cuneate neuropil with early deafferentation. A strong counter arguement
to this is that given a reduced neuropil volume on the Deaff side and
equal numbers of synapses on that side compared to the Contra side (i.e.
no effect on synapses) than equal area samples from the two sides should
result in a higher density of synapses per unit area sampled on the
Deaff side as compared with the Contra side. This was not the case.

A failure to form new afferent synapses on the Deaff side as a
result of early deafferentation might be expected to result in maintain-
ing equal density and synaptic type on the Deaff side compared with the

Contra side. However, this would necessarily assume that the inputs

36

from more central locations in the brain projecting to the cuneate
nucleus which might be forming new synapses during the time period in-
cluded in this study maintain proportionality with the Contra side.
That is, as the Contra cuneate nucleus increases in volume and adds in-
puts from the periphery, interneurons and CNS sites, the Deaff side must
be maintaining proportionality adding only synapses from interneuronal
and CNS sources, in such a fashion that the synaptic density remains
constant. Selection of survival times may again, have been inapprOpriate
to demonstrate effects of early deafferentation on synaptic formation
and maintanence. Rapid degeneration and/or withdrawal of synaptic con-
tacts present in some neonatal systems may have already occurred by 3
days of survival and the surviving synapses may occupy the maximal num-
ber of synaptic sites which remain in the smaller deafferented neuropil.
Given the lack of information concerning the effects of early deaffer-
entation on the afferent cells of the DRG which project to the cuneate
nucleus, it may be that some of these cells do survive and form fewer
synapses within the altered cuneate neuropil. Also, as the time period
of CNS input synaptogenesis is not presently known for the cuneate
nucleus, the question of whether these CNS inputs alter the density or
types of synapses present in the Deaff neuropil could be outside the
developmental period included in this study.

Finally, since these data are again pooled over survival times,
any effects present from any of the various hypotheses mentioned above
may be overwhelmed in the treatment totals. There is one alternative
hypothesis. Peusner and Morest (1977) have shown that some of the
normal developmental stages of the post-synaptic elements occur in the

absence of primary afferent innervation. Specifically, they found that

37
in the chick embryo, using Golgi LM techniques, that characteristic
spoon-type dendritic endings of the cells in nucleus tangentialis devel-
op even in the absence of primary vestibular axonal endings. They con-
cluded that some aspects of dendritic development were unaffected by
afferent input. This hypothesis can not account for the volume loss in
the cuneate nucleus due to early deafferentation. It may be that avian
development differs significantly from mammalian or these stereotyped

developmental patterns are confined to the auditory-vestibular system.
ANALYSIS II. SYNAPTIC VESICLE MORPHOLOGY
METHODS

The same micrographs used in Analysis I were used in Analysis II.
The size and shape of the synaptic sized vesicles in only those axonal
terminals identified as being pre-synaptic profiles in the sample micro-
graphs was measured directly from the micrographs using a Zeiss dissec-
tion microscope and an optical graticule. The diameter of 11 vesicles
(chosen randomly, subject to the condition that they not be overlapping)
in the pre-synaptic profiles was measured along what was visually the
long axis (L) and an axis through the widest perpendicular to the lon-
gest axis (W). Multiplying these two yields an estimate of the area of
the vesicle and dividing these two yields an index of the ellipticality
of the vesicle. The means and standard error of the means (SEM) for the
data from the individual vesicles within each terminal were then compu-
ted. It should be noted that the WxL estimate of area result in a syste-

matic over estimate of size by about one third.

RESULTS

The mean values for the width (defined as the shorter of the two
axes) multiplied by the length (wa) and the width divided by the length
(W/L) are shown for all pre-synaptic terminals in Figure 6 A for the
Contra side and Figure 6 B for the Deaff side. These are scatter plots,
intended to visually display the data. No statistical analysis of the
distributional patterns which might exist was attempted. It was felt
that a sample size of only 30 profiles per treatment would not yield
an accurate test. There does, however, appear to be a difference on a
gross level in the distributions of the synaptic vesicle means in the
pre-synaptic profiles between the two samples treatments. From Figure
6 A, showing the distribution of the Contra side terminals, there ap-
pear to be two groupings of profiles; those in the upper left quadrant
(large and relatively flat) and those in the lower right (small and
round). Figure 6 B, from the Deaff sides shows an absence of profiles
in the upper right as well as the lower right quadrants. This observa-
tion is perhaps more easily seen in Figures 7 A and B. Figure 7 A is a
frequency histogram of the ratio of W to L (W/L), arbitrarily divided
into blocks of .025 along the abscissa versus the number of pre-synap—
tic terminals whose means fall within each block, along the ordinate.
Figure 7 B is similar except the abscissa is the WxL estimate of area,
divided into blocks of 5 mm2 of enlarged area. The ordinate is the
same in Figure 7 B as in Figure 7 A. From Figure 7A, the solid bars

(Contra side) appear to be somewhat bimodal in distribution (the two

38

39

Figure 6 A. Scatter plot of mean vesicle size versus shape (Contra side).
Shown is the distribution of axonal terminal mean vesicle measurements
for size (WxL) on the ordinate versus shape (W/L) on the abscissa from
the Contra side samples. Each point represents the mean value from 11
vesicles measured within each pre-synaptic terminal computed as above.

Figure 6 B. Scatter plot of mean vesicle size versus shape (Deaff side).
Same as Figure 6 A but data is from the Deaff side samples.

3.5-1

004

15“

71%

6.54

l enlarged micrograph

2

m 0

6.0-j

le. m

55‘

56+

4.5-4

 

I
.629 .660

".4q

\

/
1/

&5‘

0.04

154

7.0-1

6.5-1

of mlorged micrograph

m

&O<

Mle In

55‘

5.0-1

#54

 

073

1a.

Contra

T
.725

Deon

40

Treatment

1
.750

Grouo

l
.775

W/L

Treatment

.030 .055

Group

.330

.eis

I
.000

 

.lll .000

Figure 6

r
.07.

r
.750

l
.775

”L

r
.000 -“l

41

Figure 7 A” Frequency of mean vesicle data for shape (W/L) of vesicles.
Shown is the number of terminals along the ordinate, whose mean vesicle
data fall within each of 9 blocks along the abscissa, representing a
range of the values of W/L (shape estimate). Each block represents a
range of .025, with black 1 containing those terminals whose mean vesi-
cle data are less than .650, block 2 terminals between .650 and .675,-
etc. Block 9 represents those terminals whose mean vesicle data are
greater than .850. Solid bars are from the Contra side data, open bars
Deaff side data, all are treatment totals.

Figure 7 B. Frequency of mean vesicle data for size (WxL) of vesicles.
Shown is the number of terminals, along the ordinate, whose mean vesicle
data for size (WxL) fall between each of 9 blocks along the abscissa,
representing a range of WxL (size estimate). Each block represents 0.50
mm of enlarged micrograph, with black 1 representing terminals whose
mean vesicle data are less than 5.5 mmz, block 2 terminals between 5.5
and 6.0 mmz, etc. Block 9 represents terminals whose mean vesicle data
are greater than 8.5 mmz. Solid bars are total treatment frequencies

of terminals in the Contra side samples; open bars those from the Deaff
samples.

42

W/L

m 8on m

use

 

 

 

 

 

 

 

 

 

1
m

1
9

.4...—«..—
876543210

$655.3 .o .3832

Blocks

Contra
Deaff
é

 

fl

 

 

       

 

5

 

4

 

 

3

 

Z

In.
r1
r1

n

 

 

4.4..1...a..
765432|O

.963th .o .3832

IC-
9
8

Blocks

Figure 7

43

modes at .725 < mean 5 i750 and .825 < mean 5 .850, with the first being
the larger). The open bars (Deaff side) might also be considered bimodal
but with the modes at .725 < mean 5 .750 and .800 < mean 5 .825, with
the latter being the primary mode. Figure 7 B, Contra side, shows a
unimodal distribution with its mode at 6.50 < mean 5 7.00. The Deaff
treatment group appears also to be unimodal (centered in the 6.00-6.50
block) but skewed to the right (representing the larger area vesicles).
Taken together, these two figures would seem to support the tentative
hypothesis that there was a higher frequency in the Deaff samples of
axonal terminals with mean size and shape estimates indicating large
and round vesicles than in the Contra samples. In addition, there
appeared to be fewer small flat vesicle containing axonal terminals on
the Deaff sides compared with the Contra sides. From Figure 7 B, one
can take the mode of each treatment side as the demarcation‘ between
large and small mean vesicle size for that treatment. Discounting those
terminals whose mean vesicle size falls within the modal block, 39% (7
of 18) of the terminals on either side of the Contra mode (6.50 < meanfé
7.00, block 5, Figure 7 B) are larger than that made. On the Deaff side,
71% (17 of 24) of the terminals had a mean vesicle size greater than the
Deaff mode (6.25 < mean 5 6.50, block 4, Figure 7 B).

There is also the trend in the Deaff treatment sides toward larger
W to L ratios, indicating more terminals on the Deaff side, compared
with the Contra treatment side with rounder vesicles (a W to L ratio of
1.00 would be indicative of a perfect round shape in a particular vesi-
cle while a mean W/L ratio for all the vesicles in an axonal terminal
profile approaching 1.00 would indicate a predominance of round vesicles).

Taking the data as bimodal in distribution for each treatment (see

44

Figure 7 A) one can sum the number of terminals within one block of the
modes for the data. On the Contra treatment side, the modes along the
the abscissa center around block 4 (mode 1) and block 8 (mode 2). 52%
(15 of 29) of all the terminals on the Contra side are within one block
of mode 1 and 26% (8 of 29) within one block of mode 2. On the Deaff
treatment side, the corresponding modes (3 and 4, respectively) are
centered in blocks 4 and 7. But these data show only 26% (8 of 31) of
all the terminals in that group having mean vesicle shape estimates
within one block of the lower, i.e. flatter mode 3 and 71% (22 of 31) of
all the terminals within one block of the higher (rounder) mode 4.

Table 4 shows the frequencies of terminals whose mean vesicle
shape (W/L) data are within one block of the modal blocks for each indi-

vidual survival time .

DISCUSSION

Any hypothesis concerning the vesicle size and shape data should
account for the two apparent trends in the data. The first trend is
for there to be more terminals on the Deaff side whose shape estimate
ratio (W/L) is larger, indicating rounder vesicles, than on the Contra
side. The other trend is an apparent increase in the Deaff treatment
samples in the number of terminals with larger mean size estimates (wa)
when compared to the Contra treatment samples. Two hypotheses appear
to be able to account for the data. One of the ultrastructural symptoms
attributed to early stage degeneration resulting from experimental
deinervation is the swelling and clumping of synaptic vesicles within

the affected terminals (e.g. Hanaway and Smith, 1979). If such a process

45

TABLE 4. FREQUENCIES OF TERMINALS WHOSE MEAN W/L RATIO IS WITHIN ONE
BLOCK OF THE MODAL BLOCK
Shown are the number of terminals within each survival time which are
within .025 of the modes in the overall data for vesicle shape (see text
for definition of modes, also Figure 7 A). ‘ . ‘MOdes.l and 2 correspond
to the two maxima in Figure 7 A for the Contra side samples, modes 3 and
4 the same maxima as in Figure 7 A for the Deaff side. Columns M91,'M,
and M+1 correspond to those blocks mode minus .025, mode, mode plus
.025. Column Sum represents the sum of the frequencies within each
surVival time and column % represents the percent of the total number of
preésynaptic terminals in each treatment which were found in the three
blocks represented in each column Sum.

Mode 1 Made 2
survival survival
time ‘M-l M M+l Sum % time M—l M M+1 Sum %
3 O 1 l 2 7 3 1 O 0 l 3
6 l 3 3 7 24 6 0 l O l 3
CONTRA
9 1 2 2 5 l7 9 0 0 0 0 O
11 0 l 0 l 3 ll 1 4 l 6 20
Mode 3 Made 4
survival survival
time M91 M M+l Sum % time M—l M M+l Sum %
3 0 2 l 3 10 3 2 2 O 4 l3
6 0 2 l 3 10 6 2 5 0 7 23
DEAFF
9 0 l 0 1 3 9 4 1 1 6 19

11 O 0 l 1 3 ll 1 2 2 5 l6

46
were occurring in the present study, one would indeed expect either
fewer terminals with smaller mean vesicle size (wa) and more terminals
with larger mean vesicle size. If the higher frequencies of large,
round terminals in the Deaff treatment group is an indication of early
degeneration, then one would expect to see some evidence of "late stage"
degeneration at the longer survival time included in this study. No
such evidence was found. It is possible that longer survival times are
necessary for the completion of the degeneration. If so, then one
would expect the frequency of large round terminals to be highest in the
longest survival period used in this study. From Table 4, one can see
that the largest number of terminals conforming to the large round type
(defined as those surrounding mode 4) were present in the 6 day survival
sample, Deaff side. This frequency declined from a high in the 6 day
sample of 7 terminals (23% of all Deaff treatment axonal terminals
measured) to 5 terminals in the 11 day survival sample (16%).

An alternative hypothesis also may account for the previously
mentioned trends in the synaptic vesicle size and shape data. CNS
axonal sprouting has been demonstrated in several systems, both develo-
ping and adult (e.g. in the developing optic system of the golden
hampster by Schneider 1873, adult dentate gyrus in the hippocampus of
adult rat by Lynch et a1. 1976 and McWilliams and Lynch 1978, septal
nuclei of the adult rat by Raisman 1969 and Raisman and Field 1973, and
in adult nucleus gracilis of the cat by Rustioni and Sotelo 1974)
Although there is some refuting evidence present in the literature that,
in adult cat there is no sprouting of the lateral funiculis afferents
to the DCN after chronic deafferentation (Rustioni and Molinaar 1975) it

nevertheless remains a possibility in the developing DCN system.

47

Jacobson (1978 pp 216) states that "in those cases where sucessful
sprouting has been demonstrated, the sprouting axons share part of the
same post-synaptic zone." He further states that the sprouting occurs
over distances of 250 m or less (Jacobson 1978, pp 211) and that the
reinnervation seen in CNS sprouting is most often "heterologous", as the
ultrastructure of the reinnervating synaptic boutons in each case is
different from the original synaptic terminals, and does not lead to
functional recovery. In-the present study, the presence of increased
numbers of terminals whose mean vesicle dimensions were large and round,
could be taken as evidence of heterologous sprouting. There are several
possible sources available for those sprouting axons which also fulfill
the other possible requirements as stated by Jacobson present in other
studies reporting CNS sprouting. ‘The requirement of proximity (250r1m
or less in distance) is certainly fulfilled by dorsal column fibers
normally innervating deeper areas of the cuneate nucleus (responding
physiologically to wrist and lower forelimb) and nucleus gracilis (re-
sponding to hindlimb and tail) are well within the 25011m limit. As
sprouting anomalous synaptic boutons rarely show functinal characteris-
tics, i.e. they show no response to stimulation which should normally
excite them, it is not necessary to assume that these boutons would
alter the receptive field characteristics of the cuneate nucleus (Hamil—
ton and Johnson 1973).

The developing CNS efferent inputs to the cuneate nucleus might
also be candidates for the origin of this hypothesized axonal sprouting.
In the absence of contrary evidence, these fibers, which certainly share
the same post-synaptic site as the DRG afferents to the cuneate nucleus

may, during the time course of this study, enter the cuneate nucleus and

48

there, upon findng "vacant synaptic sites" proliferate to a greater
extent than under normal conditions. It is also possible that inter—
neuronal connections may be present and these may proliferate as well.
It is interesting to note in this regard that while the increase in
frequency of terminals with round vesicles has reached its maximum in
the 6 day survival Deaff sample, virtually all (6 of 8) terminals with
round vesicles in the Contra treatment group were found in the 11 day
sample. Might this not be suggestive of newly arrived fiber terminals,
perhaps of CNS origin? It could be then supposed that the absence of
normal DRG terminals on the Deaff side could in some fashion, have
accelerated and altered the growth and proliferation of these arriving
terminals. Positive identification of both the time course of establish-
ment and ultrastructure of these CNS efferents in normal development is

of course necessary to address this question.

GENERAL DISCUSSION

As mentioned in the Introduction, it was expected that the area
measurements to be obtained might reflect the effects of one of several
hypotheses. These hypotheses, which might account for the nuclear
volume loss seen at the LM level after early (pre D10) deafferentation,
were: 1) degeneration of the DRG afferents; 2) cell death via transneu-
ronal degeneration; 3) abnormal cell migration; 4) atrophy and/or with-
drawal of elements of the cuneate neurons post-synaptic to DRG fibers;
5) reduced or discontinued rate of proliferation, differentation, and/or

growth of cuneate neurons.

49

1) Degeneration of DRG afferents

This hypothesis does not receive any supporting data from the
results of this study. That is, although there was slightly less area
accounted for by At profiles in the Deaff samples than in the Contra
samples, this fell short of significance (see Table 2). In addition,
only one late stage degenerating At profile was observed in each of the
treatment groups, both in the 6 day survival samples (see Figure 2).
2) Degeneratian 6f cuneate cell bodies

No cell bodies were found in the samples of either treatment group
which could be shown to have the symptoms described for degenerating
cell bodies (e.g. Chu-Wang and Oppenheim 1978a, b). That is, electron
dense cytoplasm, swollen, disrupted mitochondria, dispersion of Nissl
bodies, swelling of Golgi bodies and endoplasmic reticulum, or abnormal
chromatin and/or nucleoli. Thus, it would appear that the reaction of
the cuneate nucleus to early deafferentation does not include the trans-
neuronal deteriation of cuneate neurons, at least within the survival
times included in this study.
3) Abnormal migration

It was anticipated that evidence bearing on this hypothesis would
not be evident and none was found. No cells conforming to the charac-
teristics of migrating neuroblasts were observed in any of the samples
(c.f. Rakic 1972).
4) Dendritic atrophy

Neither degenerating Mtr profiles nor a reduced Mtr area relative
to the Contra side was found in the Deaff samples. These data would
appear to refute this hypothesis. However, the uncertainties presently
inherent in the identification of withdrawing and extending processes

(see Interpretation of profile categories Elf, Vad and Vel in Analysis

50

I Discussion) might be crucial to interpreting these data relevant to
evaluating this hypothesis. Specifically, certainly some, perhaps most
of the profiles included in the Elf and Vad categories and some in the
Vel category are dendritic in origin. The proportion of these profiles
which might be in the process of withdrawing versus that of those in the
process of extending cannot be determined by the methods used in this
study and may not be distinguishable at all. It might even be possible
that all the Elf and Vad profiles in the Deaff samples were in the
process 6f withdrawing and equal numbers of these profiles in the Contra
samples might be all in the process of extension. This would lead to a
conclusion of no difference when in fact one exists. Extension of the
survival times to be studied (both longer and shorter intervals) might
be one way to indirectly tease these two processes apart. For example,
a sufficiently long survival time might allow completion of all process
withdrawal and extension. This would then be expected to lead to differ-
ences in the final Mtr profile areas. Shorter survival times might also
lead to separating these two processes by allowing initial stages of the
processes to be observed. Earlier deafferentation might lead to the
same kind of differentiation between the two processes by allowing
withdrawal to commence at an earlier stage of development.
5) Altered development

Rate of developmental changes could not be measured in this study
because of the limited number of samples per survival time. Identifica-
tion of altered Mtr processes undergoing presumed changes is equivocal.
The arguements advanced previously for the hypothesis of dendritic
atrophy also apply to this related one, with the additional provision

that significantly more subjects per survival time be included in the

51
prOposed study.

There are several studies which seem to be suggested by the results
of the present one. Because evidence of degenerating DRG terminals was
equivocal in the present study, I would consider it extremely important
to ascertain the effects of early deafferentation upon theDRG cells
themselves. This could be accomplished using two primary measures.
Direct cell counts from DRG's (early deafferentation vessus contralat-
eral and normal controls) would establish if there were any degeneration
of DRG afferent fibers to be detected. Assuming such a cell loss, than
a reduced silver stain for degeneration could be utilized to bracket the
proper region of the dorsal column and cuneate neuropil in which to make
a subsequent search at the EM level for these degenerating profiles.

A detailed H3-thymidine experiment, initially in normal developing
subjects, later in deafferented ones, could help clear up the uncertain
effects of deafferentation on migration, if any. Initially the relative
number of cells which are still migrating during or after the period
where peripheral deafferentation results in cuneate volume loss could
be determined by methods such as those used in Schreck and Johnson (1978)
to define the dorsal a5pect of the cuneate nucleus. If cells were found
to have migrated to the cuneate after the period of deafferentation,
then the relative numbers of labelled cells which migrated to the cuneate
nucleus after deafferentation could be determined in a subsequent experi-
ment using deafferented subjects.

Effort could also be directed to resolving whether axonal sprouting
was occurring after early deafferentation and possibly the origin of
those sprouting fibers. Several techniques are available which might

provide such data. Autoradiographic injections into the nuclei of

52

origin of the CNS inputs to the cuneate nucleus would be one way to
ascertain the ultrastructure and time course of development. Lesioning
these same areas would also provide information as to the time course
of their development. Assuming that these connections developed during
the appropriate time course, EM autoradiography or EM anterograde HRP
could be employed to visualize the terminals and provide relative
frequencies, Contra versus Deaff, from which sprouting could be deter-
mined.

Another experiment using HRP, both at the LM and EM levels would
be informative in determining synaptic terminal,types in the cuneate
neurOpil. Anterograde filling of the dorsal root fibers innervating the
hand region in normal subjects (C6 through C8) could be used to determine,
at the LM level, the time course of the development and intrusion into
the cuneate neuropil of these fibers. Subsuquent EM examination of the
HRP filled terminals would allow positive identification of the DRG
input terminals and their relationship to those terminals found in the
present study to exist or remain in the Deaff sides after early deaffer-

entation.

LIST OF REFERENCES

LIST OF REFERENCES

Benes, F.M., T.N. Parks and E.W Rubel (1977) Rapid dendritic atrOphy
following deafferentation: an E.M. morphometric analysis. Brain
Res. 122, 1-13.

Brenowitz, G.L. (1980) The innervation of the glabrous forepaw skin of
developing opossums (Didelphis virginiana). in press Neurosci.

Chu-Wang, I.-W., R.W. Oppenheim (1978a) Cell death of motor neurons in
the chick embryo spinal cord. I. a light and E.M. study of natural-
ly occurring and induced cell loss during development. J. Comp.
Neurol. 177, 33-59.

(1978b) Cell death of motor neurons in the chick
embryo spinal cord. II. .a quantitative and qualitative analysis
of degeneration in the ventral root, including evidence for axonal
outgrowth and limb innervation prior to cell death. J. Comp. Neurol.
177. 59-86.

 

Cook, R.D., B. Ghetti and H.M. Wisniewski (1974) The pattern of walleri-
an degeneration in the optic nerve of newborn kittens; an Ultrastruc-
tural study. Brain Res. 75, 261-275.

Cowan, W.M. (1970) Anterograde-and retrograde transneuronal degeneration
in the central and peripheral nervous system. In Contemporary
Research Methods in Neuroanatomy (W.J.H. Nauta and S.O.E. Ebbesson,
eds.) Springer—Verlag, New York. pps. 217-251.

Del Cerro, M.P. and R.S. Snider (1972) Studies on the developing Cere-
bellum. II. the ultrastructure of the external granular layer. J.
Comp. Neurol. 144, 131-164.

Delong, G.R. and R.L. Sidman (1962) Effects of eye removal on histogene—
sis of the mouse superior colliculus: an autoradiographic analysis
with tridiated thymidine. J. Comp. Neurol. 118, 205-224.

Dvorak, A.M., M.E. Hammond, H.Fr Dvorak, and M.J. Karnovsky (1972) Loss
of cell surface material from peritoneal exudate cells associated
with lymphocyte-mediated inhibition of macrophage migration from
capillary tubes. Lab. Invest. 27, 551-574.

Falls, W. and S. Gobel (1979) Golgi and E.M. stadies of the formation
of dendritic and axonal arbors: the interneurons of the substantia
gelatinosa of Rolando in newborn kittens. J. Comp. Neurol. 187,
1-18.

53

.54

Gilbert, D. and D.J. Stelzner (1979) The development of descending and
dorsal root connections in the lumbrosacral spinal cord of the post-
natal rat. J. Comp. Neurol. 184, 821-838.

Hamilton, T.C. and J.I. Johnson (1973) Somatotopic organization related
to nuclear morphology in the cuneate-gracile complex of the opossum
(Didelphis marsupialis virginiana). Brain Res. 51, 125-140.

 

Hanaway, J. and J.M. Smith (1979) Synaptic fine structure and the termi—
nation of corticospinal fibers in the lateral basal region of the
spinal cord. J. Comp. Neurol. 1 , 471-486.

Hinds, J.W. and P.L. Hinds (1976) Synaptic formation in the mouse olfac—
tory bulb. II. Mbrphogenesis. J. Comp. Neurol. 169, 41-62.

Hughes, A. (1973) The development of dorsal root ganglia and ventral
horns in the opossum. A quantitative study. J. Embryol. Exp.
Mbrphol. 30, 359-376.

Jacobson, M. (1978) Developmental Neurobiology, 2nd ed. Plenum Press,
New‘York.

Johnson, J.I. (1977) Central nervous system of marsupials. In: The
Biology of Marsupials. (D. Hunsaker, ed.) Academic Press, New York,
pp. 159-264.

 

Juraska, J.M. and E. Fifkova (1979) An electron microscope study of the
early postnatal development of the visual cortex of the hooded rat.
J. Comp. Neurol. 183, 257-268.

Lenn, N.J. (1978a) Postnatal synaptogenesis in the rat interpeduncular
nucleus. J. Comp. Neurol. 181, 75-92.

Lenn, N.J. (1978b) Effects of neonatal deafferentation on synaptogene-
sis in the rat interpeduncular nucleus. J. Comp. Neurol. 181, 93-
116.

Levi-Mbntalcini, R. (1949) Development of the acoustico-vestibular
center in the chick embryo in the absence of the afferent root
fibers and of descending fiber tracts. J Comp. Neurol. 91, 209-
242.

Lynch, G.S., S. Mosko, T. Parks, and C.W. Cotman (1973) Relocation
and hyperdevelopment of dentate gyrus commissural system after
entorhinal lesions in immature rats. Brain Res. 50, 174-178.

Martin, G.F. and R. Dom (1970) Rubro-bulbar projections of the opossum
(Didelphis virginiana) J. Comp. Neurol. 139, 199-214.

 

Martin, G.F., J.S. King and R. Dom (1974) The organization of projec-
tion neurons in the opossum red nucleus. Brain Res. 78, l7—34.

55

Mathews, M.R., W;M. Cowan and T.P.S. Powell (1960) Transneuronal cell
degeneration in the lateral geniculate nucleus of the Macaque
Monkey. J. Anat. 94, 145-196.

Mathews, M.R. and T.P.S. Powell (1962) Transneuronal cell degeneration
in the olfactory bulb of the rabbit. J. Anat. 96, 86-102.

McWilliams, R. and G. Lynch (1978) Terminal proliferation and synapto-
genesis following partial deafferentation: the reinnervation of the
inner molecular layer of the dentate gyrus following removal of its
commissural afferents. J. Comp. Neurol. 180, 581-616.

Okado, N., S. Kakimi and T. Kojima (1979) Synaptogenesis in the cervi-
cal cord of the human embryo: sequence of synapse formation in a
spinal reflex pathway. J. Comp. Neurol. 184, 491-518.

Oppenheim, R.W., I.-W. Chu-Wang and J.L. Maderdrut (1978) Cell death of
motorneurons in the chick embryo spinal card. 111. The differenti-
ation of motorneurons prior to their induced degeneration following
limb-bud removal. J. Comp. Neurol. 177, 87—112.

Parks, T.N. (1979) Afferent influence in the development of brain stem
auditory nuclei: otocyst ablation. J. Comp. Neurol. 183, 665-678.

Peters, A., S.L. Palay and H.DeF. Webster (1976) The Fine Structure of
the Nervous System, W.B. Saunders Co. Phila. Pa.

 

 

Peusner, K.D. and D.K. Mbrest (1977) Neurogenesis in the nucleus vesti-
bularis tangentialis of the chick embryo in the absence of the pri-
mary afferent fibers. Neurosci. 2, 253-270.

Pinching, A.J. and T.P.S. Powell (1971) Ultrastructural features of

transneuronal cell degeneration in the olfactory system. J. Cell
Sci. 8, 253-287.

Privat, A., M.J. Drian and P. Mandon (1974) Synaptogenesis in the
outgrowth of rat cerebellum in organized culture. J. Comp. Neurol.

Raisman, G. (1969) Neuronal plasticity in the septal nuclei of the
adult rat. Brain Res. 14, 25-48.

Raisman, G. and P.M. Field (1973) A quantitative investigation of the
development of collateral innervation after partial deafferentation
of the septal nuclei. Brain Res. 50, 241-264.

Rakic, P. (1972) Mode of cell migration to the superficial layers of
fetal monkey neocortex. J. Comp. Neurol. 145, 61—82.

Ralston, H.J. and Ralston, D.D. (1979) The distribution of Dorsal Root
axons in Laminae I, II, and III of the Macaque spinal cord: a quan-
titative electron microscope study. J. Comp. Neurol. 184, 643-684.

56

Roscoe, J.T. (1969) Fundamental Research Statistics for the Behavioral
Sciences, Holt, Rinehart and Winston inc., New York. pp. 170-175.

Rustioni, A. and I. Molinaar (1975) Dorsal column nuclei afferents in
the lateral funiculus of the cat: distribution pattern and absence
of sprouting after chronic deafferentation. Exp. Brain Res. 23, 1-13.

Rustioni, A. and C. Sotelo (1974) Same effects of chronic deafferenta-
tion on the ultrastructure of the nucleus gracilis of the cat.
Brain Res. 73, 527-533.

Schneider, G.E. (1973) Early lesions of superior colliculus: factors
affecting the formation of abnormal retinal projections. Brain Beh.
Evol. 8, 73-109.

Schreck, J. and J.I. Johnson (1978) Postnatal days 9-15, a critical
period for the loss in cuneate nucleus due to removal of peripheral
receptive fields in pouch young opossums. Anat. Rec. 190, 534-535
(Abstract).

Singer, M., R.H. Nordlander and M. Egar (1979) Axonal guidance during
embryogenesis and regeneration in the spinal cord of the newt: the
blueprint hypothesis of neuronal pathway patterning. J. Comp.
Neural. 185, 1-22.

Skoff, R.P. and V. Hamburger (1974) Fine structure of dendritic and
axonal growth cones in embryonic chick cord. J. Comp. Neurol.
153, 107-148.

Tweedle, C.D., D.O. Hearshen and E.-M. Ostapoff (1977) Ultrastructure
of developing cuneate-gracile complex in opossums. Anat. Rec. 187,
733-734 (Abstract).

Ulinski, P.S. (1969) Normal development of the cuneate-gracile nuclear
complex in pouch young opossums. Ph.D. Dissertation Michigan State
University.

Warach, S.J. and J.I. Johnson (1977) Time course of neuronal develop-
mental stages and dendritic growth in the cuneate—gracile nuclear
region of pouch young opossums. Anat. Rec. 187, 742 (abstract).

Weibel, E. and R.P. Bolender (1974) Sterelogical techniques for elec-
tron microscopic morphometry. Ch. 6 in Principles and Techniques
of Electron MicroscOpy, vol. 3 (M.A. Hayat, ed.) Van Nostrand
Reinhold, New York. pp. 239-310.

 

West, M.J. and M. Del Cerro (1976) Early formation of synapses in the
molecular layer of the fetal rat cerebellum. J. Comp. Neurol. 165,
137-160.