Spinal cord regeneration in the shark

MISSING IMAGE

Material Information

Title:
Spinal cord regeneration in the shark
Physical Description:
x, 79 leaves : ill. ; 29 cm.
Language:
English
Creator:
Gelderd, John Bruce, 1939-
Publication Date:

Subjects

Subjects / Keywords:
Spinal Cord   ( mesh )
Regeneration   ( mesh )
Sharks   ( mesh )
Neuroscience thesis Ph.D   ( mesh )
Dissertations, Academic -- Neuroscience -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1972.
Bibliography:
Bibliography: leaves 70-78.
Statement of Responsibility:
by John Bruce Gelderd.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000897851
oclc - 25665295
notis - AEK6541
sobekcm - AA00004933_00001
System ID:
AA00004933:00001

Full Text















SPINAL CORD REGENERATION IN THE SHARK


by
JOHN BRUCE GELDERD











A DISSERTATION PRESENTED TO THE
GRADUATE COUNCIL OF THE
UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY









UNIVERSITY OF FLORIDA


1972
























DEDICATION

This dissertation is lovingly dedicated to the

memory of my father and to my mother, who so often encouraged

me to "Persevere with your studies son, you shall never

regret it."















ACKNOWLEDGEMENTS


The author takes this opportunity to thank

Dr. J. J. Bernstein for his help and advice in directing

this dissertation.

...the members of his supervisory committee for

their help and moral support.

...Mrs. Gloria Smith, Miss Suzanne Adams and Dr.

Robert Chronister for their expert histological instruction

and assistance.

...Dr. J. I. Thornby for conducting the statistical

analysis.

...Mrs. Linda Burrows for her unselfish help with

the histology, the typing and proofreading of this manuscript

and her constant moral support.

...The Department of Neuroscience and the Center

for Neurobiological Sciences for their financial support.

...and a special thanks to the people of Marineland

of Florida for the use of their facilities, their warm

hospitality and their technical assistance and encouragement.


iii





















TABLE OF CONTENTS


Acknowledgements. .


List of Tables. .


List of Figures .


Abstract. .


Introduction. .


Mammalian Studies .


Inframammalian Studie


Elasmobranch Studies.


Experimental. .


Results .. ..


Discussion .


Conclusion. .


Literature Cited. .


Biographical Sketch .


....... iii


. . v


S. vi


. . viii









.. 12


. . 21


. . 26


S . 37



S . 60


S . 68


S . 70


. . 79


1


~


1


~


~



















~~


















LIST OF TABLES


Table

1. Summary of Experimental Paradigm. ... 28

2. Summary of degenerating nerve fiber
counts in descending tracts. .. 45

3. Summary of combined left-right bouton
counts on motor horn cell bodies and
primary dendrites. .. . 47

4. Summary of strength test results
following spinal cord transaction .. 53














LIST OF FIGURES


Figure

1. Location of spinal cord sections removed
for histological analysis. . 30

2. Apparatus for testing strength of axial
musculature caudal to the site of spinal
cord transaction. .. . .. 35

3. Site of spinal cord transaction showing
cistern lined with ependymal cells at rostral
stump of cord, small cisterns within the scar
lined with endothelial cells and cisterns
in both stumps of cord, some of which enclose
severed tip of nerve fibers. . 39

A. 10 Days Postoperative ... 39

B. 30 Days Postoperative .. 39

C. 90 Days Postoperative . 39

4. A. Large severed nerve fibers showing beading
near their tips and ending in large
spherical globules. .. .. 42

B. Nerve fibers in rostral stump of spinal
cord beginning their intrusion into the
scar at 20 days postoperative . 42

C. Nerve fiber tips growing through the
scar at 30 days postoperative . 42

5. Results of bouton counts as revealed
by the Rasmussen stain ... 49








LIST OF FIGURES
(Continued)


Figure

6. A. Polygraph tracing showing undulatory
movements caudal to transaction site. 55

B. Polygraph tracing showing response of
axial musculature following stimulation
of barbels of normal animal .. 55

7. A. Results of strength tests following
spinal cord transaction. .. .. 58

B. Results of timed swimming trials
following spinal cord transaction 58

8. A. Comparison between synapse count on
motor horn cell bodies caudal to the
lesion and undulatory strength. 67

B. Comparison between synapse count on
motor horn cell dendrites caudal to the
lesion and undulatory strength. ... 67


~'c -s










Abstract of Dissertation Presented to the Graduate
Council of the University of Florida in Partial Fulfillment
of the Requirements for the Degree of Doctor of Philosophy

SPINAL CORD REGENERATION IN THE SHARK

by

John Bruce Gelderd

August, 1972

Chairman: Jerald J. Bernstein, Ph.D.
Major Department: Department of Neuroscience

The shark presents a unique central nervous system

for experimental analysis. The present series of experiments

assess the regenerative capacity of the central nervous

system in the nurse shark (Ginglymostoma cirratum).

The spinal cord was transected at the mid-thoracic

level in 28 nurse sharks. Four animals per group were

sacrificed at intervals of 10, 20, 30, 40, 60 and 90 days

postoperative. Two groups of fish underwent a subsequent spinal

cord retransection at the same site at 90 days postoperative and

were sacrificed 10 and 20 days later. Three sections of

spinal cord were removed from each shark for histological

analysis. A section of spinal cord at the site of lesion

was stained using a modified protargol silver stain to assess

regeneration across the site of lesion. Another spinal cord

section caudal to the lesion site was stained using a

viii








modified Nauta technique to show degenerating descending

nerve fibers. The Rasmussen stain was used for the light

microscopic demonstration of bouton terminaux on motor horn

cells caudal to the lesion. Behaviorally, timed trials for

swimming speed and a strength test for axial musculature

contraction caudal to the lesion site were performed at

five-day postoperative intervals.

Histological analysis showed a neuroglial-pial-

ependymal scar joining the stumps of spinal cord. Regeneration

across the site of lesion did not occur until 40 to 60 days

postoperative. Nerve fibers traversed the lesion site from

both stumps of cord and tended to follow blood vessels and

glial bridges within the scar. The number of descending long

tract nerve fibers reaching an area six spinal segments

caudal to the lesion was small (9-13%). Despite this,

synaptic terminals on motor horn cells caudal to the lesion

showed an increase from 10 to 60 days postoperative (45% of

normal at 10 days postoperative to 92% of normal at 60 days

postoperative).

Immediately upon recovery from anesthesia, all

operated sharks exhibited undulatory movements caudal to the

site of transaction while at rest which were independent of

volitional movements rostral to the lesion. These undulatory








movements increased in strength up to 60 days postoperative

at which point they were statistically indistinguishable from

the normal strength of axial musculature. Swimming prowess

was markedly reduced following spinal cord transaction and

was never recovered.. Undulatory movements were uncontrollable

and proved detrimental to swimming ability. The body caudal

to the site of lesion remained paralyzed in normal attempts

to swim.

Retransection of the spinal cord at 90 days

postoperative showed no change in the strength of axial

musculature caudal to the lesion, timed trials or the number

of boutons on motor horn cells. Comparison between the increase

in undulatory strength and increase in synaptic contacts

on the motor horn cells caudal to the lesion showed a

high correlation (r=.93, P<.01).

It was concluded that the small amount of

regeneration across the site of lesion had no effect on the

swimming behavior of the operated sharks. The increase in

strength of undulatory movements was attributed to the

reestablishment of synaptic contacts on motor horn cells

caudal to the lesion by local, segmental sprouting.















INTRODUCTION


Historically, there have been many experiments

and theories aimed at the solution to the perplexing problem

of central nervous system regeneration. One of the earliest

investigators in this area was Ramon y Cajal, who

demonstrated that growth of central nervous system fibers in

embryos resulted from protoplasmic elongation rather than

from fusion of cellular elements (Ramon y Cajal, 1928, 1960).

Ramon y Cajal's conclusions raised the question whether

similar growth would occur following a transaction of the

spinal cord of an adult animal. Subsequent studies were

undertaken in both mammalians and inframammalians to assess

anatomical and physiological regeneration of the central

nervous system.


- 1 -














MAMMALIAN STUDIES


Following the publications of Ramon y Cajal (1928)

and until the experiments of Sugar and Gerard (1940), little

was added to our knowledge of regeneration in the adult

mammalian spinal cord. During this period, numerous reports

appeared on the regeneration of fetal rat spinal cords.

These experiments met with little or no success and apparent

voluntary function was explained as spinal reflex activity.

Some abortive regeneration was seen, but the regenerated

fibers atrophied before becoming functional (Nicholas and

Hooker, 1928). In 1940, Sugar and Gerard succeeded in

obtaining structural and functional regeneration in the

transected spinal cords of young adult rats. In some of

these animals, muscle and nerve transplants had been oriented

in a longitudinal plane between the severed cord stumps.

These animals were able to sit up, clean their face,

responded by squealing when their tail was pinched and

exhibited coordinated locomotion. Stimulation of the cerebral

peduncles elicited vigorous movements of the hindquarters.

In experiments on spinal cats and dogs (Brown and

2 -






3 -

McCouch, 1947), the lack of functional regeneration in the

spinal cord was attributed to the formation of a dense

neuroglial-ependymal scar at the site of lesion. Further

study of this scarring phenomenon was done by Windle and his

associates (Windle and Chambers, 1950a, 1950b, 1951; Windle,

Clemente and Chambers, 1952; Windle, 1955, 1956; Clemente,

1955). After surgical transaction of the spinal cord of

cats and dogs, a dense neuroglial cap or scar formed over

both cut ends of the spinal cord. This scar was composed

of fibroblasts and neuroglia. The relatively avascular

cicatrix between the two stumps of spinal cord was composed

of thick masses of collagenous connective tissue and

completely isolated the cut ends of the cord. It was this

scar which was thought to prevent regeneration and thus

inhibit return of function. Administration of the

bacterially derived polysaccharide, Piromen, prevented the

development of this dense neuroglial scar and permitted the

intrusion of blood vessels into the area of lesion. As a

result of this reduction of the scar, nerve fibers were able

to cross the loose cellular matrix that formed between the

two cut ends of the spinal cord. Freeman (1955) undertook

an extensive series of experiments concerning spinal cord

regeneration in rats. His experiments lasted for more than

15 years. During this time, spinal cord transaction was






4 -

accomplished in over 7,000 rats. Anatomic regeneration and

normal functional return has occurred in approximately 100

rats. Electrophysiological studies on these rats showed

conduction of impulses in both directions through the site

of lesion. After complete functional recovery (six months

to one year), the "walking paraplegic" rats were injected

with procaine directly into the spinal cord at the site of

transaction. The animal again became completely paraplegic.

Retransection of the spinal cord in such rats also returned

them to a paraplegic state. Freeman (1955), Littrell et al.

(1953) and Littrell (1955) transected the spinal cord in

adult cats and contrasted the recovery of non-treated animals

to those treated by intravenous injections of Piromen. Cats

treated with Piromen exhibited return of function beginning

at 2-3 months postoperative and peaked at 9-12 months. After

this point, the animals regressed to typical paraplegic

behavior by 18 months. Histology showed a dense neuroglial-

ependymal scar at the site of transaction which was "choking

off" the regenerated fiber tracts. There was never any

return of sensory function. Non-treated animals did not

show any restitution of function or anatomical regeneration.

These studies indicated that the neuroglial-ependymal scar

prevented regeneration in mammals. However, this same

scarring phenomenon also occurred in inframammalian forms that






5 -

did indeed show both anatomical regeneration and

physiological return of function after spinal cord

transaction. It was thought, therefore, that perhaps

mammalian central nervous system neurons lacked sufficient

growth potential to. regenerate lost peripheral processes.

Levi-Montalcini and Brooks (1960) tested the effects of a

protein isolated from the mouse salivary gland upon chick

and mouse sensory ganglia in vitro. Within 12 hours after

injection of this nerve growth factor (NGF) into the medium,

a dense halo of nerve fibers surrounded the explant.

Injection of NGF into intact animals produced hyperplasia

and hypertrophy of sensory and sympathetic neurons. Scott

and Liu (1964) injected NGF and Piromen into kittens after

crushing the dorsal columns. There appeared to be a

definite, direct correlation between the amount and duration

of NGF administration and the regeneration of the sensory

fibers. Anatomical regeneration across the site of lesion

was confirmed electrophysiologically in these kittens

although Scott and Liu did not wait for functional return.

Scott et al. (1966) administered NGF to young rats after dorsal

root crushing and found an increase of 14% in protein production

in dorsal root ganglia as compared to no increase without

administration of NGF. Harvey and Srebnik (1967) found

anatomical and physiological regeneration with return of function





6 -

following spinal cord compression in rats treated with

L-thyroxine. Non-treated rats showed no regeneration or

return of function. In another rare case of central nervous

system regeneration, Adams et al. (1968, 1969, 1971) cut the

infundibular stalk in ferrets. Degeneration occurred

initially, followed within two weeks by new fibers

regenerating from the hypothalamus through the fibrous scar.

These fibers carried neurosecretory material. During the

one-to three-month postoperative period, fibers grew to a

proximal ectopic infundibular process which formed following

the lesion. In animals kept alive until 12 months

postoperative, the entire neurohypophysis had been

reinnervated and was functional.

If a nerve fiber is severed, there appear to be at

least three distinct reactions: (1) growth does not occur

or it is abortive; (2) the original fiber may regenerate

and reform its synaptic contact or (3) adjacent intact

nerve fibers or cells may develop collateral sprouts and

reinnervate the deafferented tissue (Guth and Windle, 1970).

Liu and Chambers (1958) experimentally demonstrated sprouting

in the spinal cord of the cat following deafferentation

either by adjacent dorsal root section or by corticospinal

tract ablation at cranial levels. They observed that

paraterminal and collateral sprouting was rather generalized






7 -

in areas which had been deafferented and that the amount

of this sprouting was determined by the extent of the

denervation. Goodman and Horel (1966) showed restricted

sprouting of optic tract projections in the rat after

occipital cortex removal. Schneider (1970) lesioned the

visual cortex or superior colliculus in neonate and adult

hamsters. The neonate hamsters showed sprouting in optic

tract projections and some sparing in visual discrimination

tests in contrast to the adult hamsters which showed little

or no anatomical regeneration or return of function. By

using histochemical fluorescence techniques, Bjorklund and

his colleagues have shown regenerative axon sprouting in

the rat mesencephalon following electrolytic lesions

(Katzman et al., 1971; Bjorklund and Stenevi, 1971) and in

the rat spinal cord following spinal cord compression

(Bjorklund et al., 1971).

Raisman's investigations (1966, 1969a, 1969b)

in the septal nuclei of the adult rat constitute one of the

few ultrastructural studies on collateral sprouting in the

mammalian central nervous system. Afferents from two

separate pathways converge upon the medial septal nucleus.

Fibers originating in the hippocampus pass to the septum

through the medial forebrain bundle. Hippocampal fibers

terminate exclusively on the dendrites of the septal nuclei





-8-

whereas the hypothalamic fibers terminate primarily upon the

cell bodies of the septal nuclei. After lesioning the

hippocampal input, the remaining afferent septal fiber tract

showed a high proportion of axon terminals which made contact

with more than their normal share of postsynaptic units.

This phenomenon was interpreted as a reinnervation by the

remaining hypothalamic input of synaptic sites left open by

lesioning of the hippocampal input. This was corroborated

by lesioning the medial forebrain bundle which produced

degeneration of the remaining synaptic contacts. Raisman

then lesioned the hypothalamic input and found sprouting

of hippocampal fibers in the septum to fill synaptic sites

left by the degenerated hypothalamic input. Moore et al.

(1971) has recently duplicated Raisman's study utilizing

histochemical fluorescence techniques and has corroborated

Raisman's findings. Ultrastructural experiments on rat

spinal cord (Bernstein and Bernstein, 1971) have shown

similar results. Neurons were deafferented by hemisecting

the spinal cord. Shortly after these cells were deafferented,

they began to hypertrophy. In particular, the dendrites

exhibited profuse branching and irregular swelling. Synaptic

spines on the hypertrophied dendrites were also greatly

increased in number. Concomitant with this was the

establishment of large numbers of axodendritic synapses.






9 -

Furthermore, the axons that terminated near the site of

lesion appeared to arise by way of axonal sprouting from

the region of the spinal cord that was not hemisected. To

ascertain if there was any long tract involvement in the

regenerative process, a group of rats underwent hemisection

of the spinal cord just below T2. Ninety days later, the

spinal cord was hemisected on the same side at C5 and

the spinal cord at vertebral level T2 underwent histological

examination for degeneration. The region immediately rostral

to the T2 lesion showed new, degenerating nerve fibers in

small amounts. The area was also filled with degenerating

axodendritic synapses. These data suggested that the long

tracts were indeed partially involved in the regeneration of

axons to the neurons proximal to the site of the original

lesion. Bernstein stressed, however, that the number of

fibers appeared to be low and that the vast majority of new

synapses originated from segmental sprouting. This study

showed, however, that descending long tract nerve fibers in

the spinal cord of the rat are capable of limited

regeneration to the area immediately rostral to the lesion

site. The regenerating axons in the rat appeared to respond

to nonspecific influences of the hypertrophied dendrites

and established inappropriate connections. The formation

of these synapses then effectively terminated further growth






10 -

of the axons. Bernstein and Bernstein (in press) have also

shown limited regeneration of axons rostral to the site of

hemisection in the Rhesus monkey spinal cord. Motor horn

cell dendrites immediately rostral to the site of

hemisection showed varicosities. Regenerating axons made

normal as well as aberrant synaptic recombinations with

reactive neurons rostral to the lesion. The most frequent

type of aberrant synaptic complex was a cup-shaped bouton with

a central, large extracellular space between presynaptic and

postsynaptic membranes. In another recent ultrastructural

study, Lund and Lund (1971) found synaptic adjustment in the

superior colliculus following enucleation of neonatal and

adult rats. Little change was observed in the number and

types of synapses in neonatal rats due to synaptogenesis,

but adult rats showed a reinvasion of synaptic sites with

an incomplete return to a normal proportion of synaptic

types.

Although regeneration in the mammalian central

nervous system appears possible, functional return has been

shown in only isolated cases. The majority of experiments

have shown only abortive regeneration. In the instances of

functional return after spinal cord lesion, many are regressive

and the animal returns to a paraplegic state. Sprouting then

appears to be the rule in the mammalian central nervous system,






11 -

with functional regeneration the rare exception. There is,

as yet, no clear functional significance to the phenomenon

of central nervous system sprouting. McCouch et al. (1955)

has implicated sprouting in spasticity following spinal

cord transaction and Schneider (1970) suggests that sparing

of pattern vision behavior in neonatal hamsters is due to

sprouting of visual pathways. However, this evidence is

only suggestive at best and more definitive studies are

required to shed more light on this anatomical phenomenon.















INFRAMAMMALIAN STUDIES


In contrast to the abortive regeneration found in

mammals, inframammalian forms have proven to be a fertile

area for successful central nervous system regeneration

studies. In fact, central nervous system regeneration seems

to be the rule in lower forms with abortive regeneration

the exception. Central nervous system regeneration in bird

embryos has been well documented (Clemente, 1955; Hamburger,

1955) although regeneration in the adult bird occurred only

in the visual system (Cattaneo, 1923). Central nervous

system regeneration studies in reptiles have centered

around the well known phenomenon of tail regeneration in

lizards (Clemente, 1964; Hamburger, 1955). The central

nervous system of amphibia has shown great regenerative

powers. With the noted exception of the adult Anurans, the

amphibia have proven comparable to teleosts in central nervous

system regeneration. After complete transaction of the spinal

cord of larval salamanders (Piatt, 1955a, 1955b) or in the

axolotl (Kirsche, 1956), extensive regeneration of fiber

tracts was observed with a concomitant return of normal

12 -






13 -

function. In the adult salamander, regeneration was equally

vigorous (Piatt, 1955a, 1955b). Regeneration of the spinal

cord of the adult newt took place within 30 days with or

without injection of Piromen (Drummond, 1954). Regeneration

in the frog central nervous system has been restricted

to the larval stages (Hooker, 1925). The adult frog has

shown abortive regeneration of the spinal cord after

transaction (Piatt and Piatt, 1958; Clemente, 1964).

Regeneration in the frog and toad visual system, however,

has been very specific and successful, both anatomically

and physiologically (Sperry, 1944; Gaze and Jacobson, 1963;

Gaze and Keating, 1969, 1970a, 1970b; Gaze, 1970).

The regenerative capacity of the fish spinal cord

ranks high among the vertebrates. This has been shown

repeatedly through the almost exclusive use of the teleost

as an experimental animal in regeneration studies.

Regeneration in cyclostomes has been largely restricted to

spinal cord regeneration in larval forms (Maron, 1959;

Hibbard, 1963; Niazi, 1963). Spinal cord regeneration in

teleosts has proven so superior to the mammalian nervous

system that what would be considered poor functional or

anatomical recovery in teleosts would undoubtedly be hailed

as strikingly successful in man or any commonly used

laboratory mammal. Regeneration in teleosts does not occur






14 -

to the same degree in all parts of the central nervous

system. Regeneration throughout the central nervous system

in fish has recently been thoroughly reported by Segaar

(1965) and Bernstein (1970) and will not be repeated here.

The discussion here will be largely restricted to spinal

cord regeneration.

The regenerative capacity of the fish spinal

cord extends from the simple regrowth of axons across the

site of lesion to the complete restitution of neural

cytoarchitectonics, replete with new nerve cells and glia.

Koppanyi and Weiss (1922) carried out spinal cord transactions

at a high level in goldfish. The fish were paralyzed caudal

to the lesion for two to three weeks, after which they began

to show signs of return of function. After 60 days, the fish

were behaviorally indistinguishable from normals. Histological

examination showed regeneration of neural pathways which

resulted in the reappearance of normal connections (Koppanyi

and Weiss, 1922; Koppanyi, 1955). Pearcy and Koppanyi (1924)

later cut the entire vertebral column of goldfish with

scissors so that no bony continuity remained between the

regions anterior and posterior to the section. Ten weeks

postoperatively, the fish were again swimming normally.

Hooker (1930, 1932) transected the spinal cord of guppies

less than four days old. He claimed full coordination and





15 -

integrative movements concomitant with the reestablishment

of nervous connections between the two halves of the body

approximately four days postoperatively. Keil (1940)

transected the spinal cord of adult rainbow fish and claimed

restitution of function beginning from three to twelve days

postoperative with complete restitution of function at

30-40 days postoperative. Ten years later, Kirsche (1950)

confirmed not only the functional but also the morphological

regeneration of the spinal cord in the adult rainbow fish.

Kirsche introduced the method of stimulating the spinal cord

above the site of transaction. Tail fin movements were

elicited when the cord, and only the cord was stimulated

above the transaction. Those animals which showed no

morphological regeneration also showed no movement of the

caudal fin upon stimulation of the spinal cord above the

transaction. Kirsche distinguished various phases during

the course of regeneration. The first phase, which was

apparent approximately four days after the transaction,

consisted of a disorganized growth from the severed stumps.

The second phase began about seven days postoperative with

a mitotic increase of the ependymal cells to form "indifferent

neural cells" which in time developed into neuroblasts and

glioblasts. Further differentiation lead to the formation

of normal cells in both proximal and distal stumps. Oriented





16 -

fibers grew out from these new cells and,approximately 15

days postoperatively, there was evidence for both morphological

continuity and functional recovery (Kirsche, 1950, 1965).

Healey (1962) transected the spinal cord of minnows and noted

the immediate inability of the minnow to change colors upon

background color reversal. Fast color changes were shown to

be under autonomic control and slow color changes were

mediated by hormonal control. Ten days after transaction,

the fishes ability to change color increased. After four

months, rapid color changes occurred that were indistinguishable

from normal. Bernstein (1964) has shown a relationship

between age and regenerative capacity of the goldfish spinal

cord. Young goldfish (less than one year old) were able

to reconstitute approximately 90% of the available descending

axons whereas approximately 60% were reconstituted in two-

and three-year-old animals. The ability of the neuroglia

to regenerate and reestablish the diameter of the cord was

also age dependent. The younger goldfish reconstituted the

diameter of the cord almost completely (Bernstein, 1964).

Not only has the spinal cord of the teleost fish regenerated

after being severed, it also has the ability to completely

reconstitute areas of the spinal cord following ablation.

This type of growth pattern has been found in the regeneration

of the caudal neurosecretory system of Tilapia. After removal






17 -

of the caudal peduncle, tailfin and caudal spinal cord

segments, a new caudal neurosecretory system regenerated.

This system was somewhat abberant but fully functional

(Fridberg et al., 1966).

In a series of experiments on goldfish, it has been

shown that although there was return of function following

spinal cord transaction, the morphological regeneration was

less than perfect (Bernstein, 1964; Bernstein and Bernstein,

1968; Bernstein and Gelderd, 1970; Bernstein and Gelderd,

in manuscript). Following spinal cord transaction, goldfish

were paralyzed caudal to the site of lesion, descending spinal

tracts degenerated, and synaptic sites on perikaryon and

primary dendrites of motor horn cells 2 cm caudal to the site

of lesion dropped by 50%. Following 60 days regeneration

time, the synaptic complement on motor horn cells was

reestablished although descending fiber tract regeneration

was only 35-50% of normal. A subsequent retransection of the

spinal cord at 60 days postoperative resulted in degeneration

of the new, regenerated descending fibers and concomitant

paralysis caudal to the site of lesion. In contrast, there

was no statistically significant change in the synaptic

complement on motor horn cells 2cm caudal to the site of

lesion. This seemed to indicate that the descending fibers

regenerating into the caudal section of spinal cord did not






18 -

return to their original synaptic sites on motor horn cell

perikaryon or primary dendrites, but perhaps synapsed instead

on internuncial cells. Return of the normal synaptic complement

on motor horn cell perikaryon or primary dendrites was

relegated to local, segmental sprouting of adjacent fiber

tracts or cells. Those descending fibers which did not

regenerate past the site of lesion appeared to synapse on

cells near the lesion site. When the regenerating axons

reached the site of lesion, they were confronted with

deafferented neurons. The regenerating axons synapsed on

these cells until the synaptic sites were filled. Once the

regenerating axons made these inappropriate synaptic contacts,

they ceased their growth. The mechanism for the cessation

of growth of these axons is thought to be a special case of

contact inhibition (Bernstein and Bernstein, 1968). It must

be stressed again that regeneration in the goldfish central

nervous system is not a uniform phenomenon. Although

transaction of the spinal cord was followed by a reduced

regenerative capacity in the number of fibers traversing the

lesion (Bernstein and Gelderd, 1970), lesions of the visual

system were followed by a more specific regenerative process

which initially appeared to be point for point with 100% of

the original optic fibers regenerating (Attardi and Sperry,

1963; Jacobson and Gaze, 1965). Recent studies, however,






19 -

have shown some deviation in the area of termination of

regenerating optic nerve fibers in the optic tectum of the

goldfish following lesions in the visual system (Yoon, 1971;

Horder, 1971; Sharma, 1972).

In mammalians, the neuroglial-ependymal scar is

thought to be responsible in part for the apparent lack of

regeneration in the central nervous system, forming a dense

barrier between the cut ends of the spinal cord and thus

preventing regeneration (Brown and McCouch, 1947; Windle,

1955; Windle and Chambers, 1950a, 1950b; Guth and Windle,

1970). Bernstein and Bernstein (1967) investigated the

effect of the neuroglial-ependymal scar on spinal cord

regeneration in goldfish. The spinal cord was completely

transected and a thin teflon disc was placed between the two

cut ends of the spinal cord. The teflon disc remained

between the cut ends for 30 days which was ample time for

regeneration to occur in goldfish. The goldfish were then

operated upon again and the teflon disc removed. These

goldfish were observed for an extended period of time to

determine any return of function caudal to the lesion which

would signify regeneration of the spinal motor tracts. No

regeneration or return of function was observed six months

later. Other goldfish were operated upon again at 30 days

postoperative. The teflon disc was removed and the spinal






20 -

cord transected one spinal segment rostral to the original

lesion. After 30 days, the spinal cord was observed

histologically to assess regeneration. The descending

spinal motor tracts grew through the second lesion, caudalward

through the isolated section of the spinal cord and through

the first lesion which was delineated by a substantial

neuroglial-ependymal scar. Hence, the regenerative capacity

of the goldfish spinal cord was not affected by the

neuroglial-ependymal scar acting as a mechanical barrier

(Bernstein and Bernstein, 1967).















ELASMOBRANCH STUDIES


The shark presents a unique nervous system, both

among the fishes specifically and vertebrates in general.

One trait which is particularly unique to the elasmobranchs

is the reported absence of internuncial cells (Golgi Type II)

in the spinal cord (Kappers et al., 1936; Von Lenhossek,

1892, 1895; Aronson, 1963; Nieuwenhuys, 1964). Presumably,

the vast majority of descending tracts end directly on

motor horn cells without intermediary neurons to intercede or

modulate information from higher centers. Another unique

characteristic restricted to the elasmobranchs is the high

urea content in the blood and sera (Goldstein, 1967; Goldstein

et al., 1968; Rasmussen and Rasmussen, 1967; Smith, 1929;

Rasmussen, 1971). This phenomenon is particularly evident

in marine elasmobranchs. This large concentration of urea

is also reflected in the cerebrospinal fluid (Smith, 1929;

Rasmussen, 1971) and may have a profound effect on the

regenerative capacity of the shark central nervous system

because urea is used to shrink brain tissue during neurological

operations.


- 21 -






22 -

In addition to the above-mentioned characteristics

of the shark central nervous system, this class of fish

shows important advancements in the evolution of the

vertebrate central nervous system. The structure of the

spinal cord in elasmobranchs may be considered a prototype

for that of high vertebrates. Unlike the spinal cord of

cyclostomes which is flat, the spinal cord of elasmobranchs

is round or oval. In addition, these are the first primitive

animals to have myelinated fibers in the spinal cord and

whose dorsal and ventral roots unite outside the vertebral

column to form a mixed root. It is also in this class that

one first finds the division of the gray matter into dorsal

and ventral horns and the first time that all cells of

origin for sensory fibers in the cord lie in extramedullary

spinal ganglia (Kappers, 1906; Kappers et al., 1936; Aronson,

1963; Nieuwenhuys, 1964).

In mammals and other inframammalian forms,

transaction of the spinal cord produces paralysis immediately

after transaction. In most inframammalians, paralysis is

alleviated by regeneration and return of function after

varying periods of time. In the case of mammals, spinal

walking may occur after weeks or months of recuperation.

The shark, however, has exhibited coordinated undulatory

movements immediately upon recovery from anesthesia after






23 -

spinal cord transaction (Ten Cate and Ten Cate-Kazejawa,

1933; Gray and Sand, 1936a, 1936b; Lissmann, 1946a, 1946b;

Healey, 1957) and was able to swim, using coordinated

movements between those portions of the body rostral and caudal

to the transaction. Ten Cate and Ten Cate-Kazejawa (1933)

removed all the muscles in the region of the anterior dorsal

fin of the dogfish (Scyllium canicula and S. catalus) and

transected the spinal cord in the same region. He observed a

locomotory rhythm propogated over the site of the operation,

thereby maintaining coordinated movement between the head and

posterior region of the body. According to Ten Cate and Ten

Cate-Kazejawa (1933), the activity of the posterior region

of the body depended upon tensile stimuli applied to the

posterior musculature whenever an active contraction occurred

in the head region. If this conclusion is justified, then

the normal locomotory rhythm of the dogfish involves the

activity of a chain of peripherally controlled reflexes.

Gray (1936) and Gray and Sand (1936a, 1936b) disagreed with

the hypothesis of Ten Cate and Ten Cate-Kazejawa. They

showed that coordinated responses no longer occurred if two

regions of the body of a dogfish were isolated from one

another by a second spinal cord transaction. If both of

these regions were of sufficient length, each exhibited an

independent, spontaneous, automatic activity within the






24 -

spinal cord. Lissman (1946a, 1946b) showed that locomotory

rhythm could only be abolished in spinal dogfish by a

complete, bilateral rhyzotomy caudal to the transaction.

This conclusion appeared to support Ten Gate's hypothesis.

The continued locomotory rhythm after spinal cord transaction

in sharks may in part be attributed to its rather low

position on the phylogenetic ladder. Eels have shown an

undulatory behavior after decapitation, having a duration of

only a few seconds (Gray, 1936). Gray has attributed this

phenomenon to injury potentials in the remaining spinal

cord. Typically, spinal eels laid on their side and showed

no coordination between rostral and caudal portions of the

body in normal attempts to swim. Nociceptive stimuli did,

however, cause undulations (Gray, 1936). Little mention has

been made of swimming prowess in the shark with respect to

strength and speed after spinal cord transaction. An attempt

to clarify these questions was made during this series of

experiments.

Central nervous system regeneration studies

in elasmobranchs are conspicuous by their absence. Although

lesion studies have been done (Ten Cate and Ten Cate-

Kazejawa, 1933; Gray and Sand, 1936a, 1936b; Lissman,

1946a, 1946b; Healey, 1957; Segaar, 1965; Ebbesson, 1972),

there have been no known experiments concerning






25 -

shark central nervous system regeneration. Indeed, experimental

work per se on the shark central nervous system is at best

limited. It is this complete lack of central nervous system

regeneration experiments on sharks plus the anatomical and

behavioral uniqueness of their central nervous system that

have prompted me to use them as experimental animals. It was

my intent to use these unique features of the shark central

nervous system to shed some light on the regenerative

process.















EXPERIMENTAL

Materials and Methods


Subjects

Thirty-six male and female nurse sharks

(Ginglymostoma cirratum), approximately two feet in length,

were used. These sharks were trapped in the coastal waters

off Ft. Lauderdale, Florida.

Environment

All fish were kept at Marineland of Florida in an

outside, circular, salt water tank 15 feet in diameter and

six feet in depth. A constantly circulating salt water

system was used to insure proper oxygenation, salinity and

water temperature. Sharks were fed to satiation daily on

cut-up fish.

Operative Procedures

All operated fish were anesthesized with Tricaine

methanesulfonate (MS-222, 1:4000, Finquel, Ayerst Laboratories),

then placed on an operating board. A longitudinal incision

was made at the midline in the midthoracic region at the level

of the trailing edge of the pectoral fin, and the musculature

26 -





27 -

dissected away to expose the spinal column. A laminectomy

was performed and the spinal cord transected with a scalpel.

The wound was sutured and powdered sulfathiazole-sulfonilimide

was applied to the suture line to prevent infection. Following

surgery, all animals were also given a 0.1 cc intramuscular

injection of Longicil. Animals were tagged for identification

by attaching a numbered clamp and colored streamers to the

anterior dorsal fin. All sharks, including four normals,

were separated into nine groups (four animals per group)

and were killed by anesthetizing them at 10, 20, 30, 40, 60

and 90 days postoperative and perfusing them with 10% buffered

formalin. Two groups of animals underwent a subsequent

retransection of the spinal cord at the same site at 90 days

postoperative and one group each was killed by the above

fixation method at 10 and 20 days postoperative (Table 1).

Histology

Three sections of the spinal cord were removed

from each shark: a 2 cm section at the site of lesion, a

second section (1 cm in length) six spinal segments caudal

to the site of lesion and a 1 cm section immediately caudal

to the second section (Fig. 1).

The 2 cm section of spinal cord at the site of

lesion was sectioned horizontally at 15q and stained using

a modified Bodian silver technique counterstained with










U) > H r-I rd
W) (d i r-l r0
E-4 Q > rd


S-1 r. 4-)









__---- -+ -
> >M 0
4 C t






H (- m N e
r) X X X X X X X X X0



69 P C 0 o ,c
4J
rl o

19 -- ___ __ __ __ __ -- -- -- f tn


S0r 0

1 r4 r-1-
W 1 C. 0 0
: M (0 x X x x x X X
F:0 4-) 9 0





H En ) >-
H Z H r 0d





0
Q ..o ..... l u
d b





H v dnI
1 ni r1 44C(
z o a) Q)



r-4
w r r24 0 0






0 O I
0 r4 pq 3 > z r- c
S- H s +rrj




O M4 4p
*(d 4 1 U
w Q 0 r q I
PIC i -






H0) 00 0 0 0 0 0 0 0 0)
S-H r-4 r- )
p3 0 (I 0 r0
() N (n F:-



0 r *i











O0 c C-S
r p z o 0) o W a) a0)
ri 0 (a >4 >4 > > > >q 1 0 >4 *

H *H1 C







w- H Na 0
00 O O
0 zI< Wt U) Ud
z C IV. 01 pt



0o e p N m It n ko 00 a% Q
O .rl .r .o 0

I= I-1-1


* uf























Figure 1 Location of spinal cord sections removed for
histological analysis.







- 30 -


Site of
Spinal Cord Transection
Section Taken
for Bodian Stain


Spinal Cord
Section Taken
for Nauta Stain


Spinal Cord Section
Taken for Rasmussen
Stain


segments






31 -

cresyl-violet and eosin. The regenerative process was

assessed with respect to (1) formation of a neuroglial-

ependymal scar, (2) the effect of the scar on regeneration

and (3) the rate of the regenerative process.

Three descending tracts (tectospinal, thalamospinal

and ventral cerebellospinal tract) were studied in the 1 cm

segment of spinal cord located six spinal segments caudal

to the site of lesion. This section of spinal cord was

histologically analyzed with respect to the number of

degenerating nerve fibers within a given tract following

spinal cord transaction (Fig. 1). The spinal cord was

serially sectioned horizontally at 30A and impregnated

using a modified Nauta technique for degenerating nerve

fibers. The degenerative pattern was plotted by drawing a

composite cross-sectional diagram made by examining, in order,

each horizontal section. The histological sections

corresponding to the known anatomical locations of the

given descending tracts were selected and the number of

degenerating fibers counted in each tract at 10 and 20 days

postoperative. These data were compared to the number of

degenerating nerve fibers found 10 and 20 days following

the subsequent retransection (Table 1). The second transaction

controlled for local versus long tract input following

regeneration.





32 -

A quantitative analysis of the synaptic terminals

was done on the perikaryon and primary dendrites of ventral

motor horn cells following spinal cord transaction. The

section (1 cm in length) of spinal cord used was taken

immediately caudal to the section used for fiber tract

analysis so that the number of regenerating fibers and the

synaptic profiles could be compared (Fig. 1). The spinal

cord was sectioned coronally at 10l and impregnated using

the Rasmussen stain for the light microscopic demonstration

of bouton terminaux (Rasmussen, 1957) and followed by a

cresyl-violet and eosin counterstain. Counts were made on

only those motor horn cells in which a prominent nucleolus

and primary dendrite could be seen in a given section.

Counts were made on a total of 576 motor horn cells. Sixteen

motor horn cells were counted per shark, utilizing eight

cells on the left side and eight cells on the right side.

Synaptic terminals were counted at 10, 20, 30, 40, 60, and 90

days postoperative and on the two groups of fish retransected

at 90 days postoperative (Table 1). All counts were made on

coded slides to insure unbiased results. The resultant data

were decoded and the levels of significance determined for

intra- and intergroup interactions by using computer

program BMDX63 for multivariate analysis of variance.





- 33 -


Behavior

A behavioral analysis of the sharks was done

during the postoperative period including the two groups

retransected at 90 days. The operated sharks were observed

daily while swimming in the tank and compared to normal

sharks with respect to swimming prowess. In addition, two

quantitative tests were performed on all sharks,

preoperatively and at five-day postoperative intervals.

The first test consisted of removing each shark from the

tank and strapping it to a board with that portion of the

body rostral to the lesion firmly held in place. That

portion of the body caudal to the lesion remained unrestrained

with the exception of the caudal peduncle to which a hose

clamp was attached. The clamp was connected by way of a

#10 screw to a Statham load cell assembly (Model UL-4)

which was in turn mounted on a Statham universal force

transducer (Model UC-3). The entire transducer assembly

was securely mounted on the test board. The output of the

force transducer was fed into a two-channel Grass polygraph

recorder. Two electrodes were attached to the paired

barbels located on the underside of the snout of the shark

(Fig. 2). The shark was stimulated using a constant current

stimulator producing a 10 ma pulse of 50 msec duration.

The strength of the response of the caudal body musculature
























Figure 2 Apparatus for testing strength of axial
musculature caudal to the site of spinal
cord transaction.






- 35 -


----CONSTANT CURRENT
SSTIMULATOR
-'V2


ELECTRODE ATTACHED
TO BARBEL


-RESTRAINING
STRAP




TRANSACTIONN SITE


LOAD CELL ASSEMBLY






36 -

following stimulation was recorded on the polygraph and

compared to preoperative and normal data. A minimum of

five responses was recorded for each shark on a given

trial day.

The second test consisted of timed swimming

trials. Each shark was placed in the water at one end of a

7' x 3' x 2' tank and the time required for the shark to

swim the length of the tank was recorded. Two consecutive

timed trials were measured on each shark on a given trial

day. Only those trials were counted in which swimming was

uninterrupted over the entire distance. The postoperative

data for both behavioral tests were compared to normal data

in addition to intra- and intergroup interactions among

postoperative groups by utilizing computer program BMDO8V

for analysis of variance.















RESULTS

Histological


Site of Lesion

Histological analysis at the site of lesion in

the group of animals sacrificed at 10 days postoperative

showed a dense scar separating the cut ends of the spinal

cord (Fig. 3A). The diameter of the cord in the scar area

was approximately 75% of normal. The scar was composed of

neuroglial,ependymal and pial cells. Blood cells and

phagocytes were also present in abundance throughout the

scar, but no blood vessels were seen within the scar at

this time. Three types of cisterns were found within or

near the site of lesion. A cistern lined with ependymal

cells was visible at the rostral stump of spinal cord.

Several smaller cisterns lined with endothelial cells

were also present throughout the scar. A third type of

space or cistern was found near the lesion site in both

stumps of spinal cord. These spaces were not lined with

cells and some had severed nerve fiber tips within them.

All three types of cisterns were seen in the lesion

-,37 -























Figure 3 -


Site of spinal cord transaction showing cistern
lined with ependymal cells ((), at rostral
stump of cord, small cisterns within the scar
(') lined with endothelial cells and cisterns
in both stumps of spinal cord ( ) some of which
enclose severed tips of nerve fibers. Bodian
silver stain (X10).

A. 10 days postoperative

B. 30 days postoperative

C. 90 days postoperative cut ventral root
growing from rostral stump of cord into
scar area (f).





- 39 -


A











B











C


,Cl~g+HI-~
Il 71~a





40 -

area during the entire postoperative period (Fig. 3A, 3B,

3C). No nerve fibers were found within the scar at this

time. Many large nerve fiber tips were,in fact, found well

back from both cut ends of spinal cord (Fig. 3A). These

large nerve fibers were beaded at their terminals and ended

in large spherical globules (Fig. 4A). This phenomenon

persisted throughout the postoperative period.

At 20 days postoperative, the scar area resembled

that at ten days. Nerve fibers were seen immediately adjacent

to the scar with some fibers beginning their intrusion into

the scar from both stumps of cord (Fig. 4B).

At 30 days postoperative the cistern lined with

ependymal cells had increased considerably in size and the

diameter of the scar was further reduced (55% of normal).

Nerve fibers were seen penetrating the scar from both

stumps of cord (Fig. 4C) but the center of the scar was

still devoid of nerve fibers.

At 40 days postoperative there was little change

in the appearance of the scar or the density and intrusion

of nerve fibers into the scar area.

There were no discernable differences in the

appearance of the lesion site between 60 and 90 days

postoperative. The scar appeared as a loose cellular matrix

made up of neuroglial and ependymal cells. The ependyma-






















Figure 4 A. Large severed nerve fibers showing beading
near their tips and ending in large spherical
globules ( ). Bodian silver stain (X1,000).

B. Nerve fibers in rostral stump (RS) of spinal
cord beginning their intrusion into the scar
(S) at 20 days postoperative. Bodian silver
stain (X450).

C. Nerve fiber tips growing through the scar at
30 days postoperative. Bodian silver stain
(x1,000).





- 42 -


mow,

pL~DY

* ) 4 I r ; L l


S 1


* 4
'.1

*#r

ArJ


-i





43 -

lined cistern at the rostral stump of cord had increased

still more in size (Fig. 3C) but no further constriction

in the diameter of the scar was observed. The increase in

the size of the ependyma-lined cistern in the rostral stump

of spinal cord was probably due to blockage of the flow of

cerebrospinal fluid in the central canal of the spinal cord.

There was a large increase in the number of nerve fibers

within the site of lesion. These nerve fibers were of

small caliber and appeared to completely traverse the scar

by following neuroglial bridges and blood vessels. In those

sharks where the transaction was made immediately caudal to

the emergence of a pair of ventral roots, the ventral roots

were severed in the process of transecting the cord.

Severed ventral roots were seen, at 60 and 90 days

postoperative, growing caudally along the edge of the cord

until they reached the lesion site where they grew into the

scar and the caudal stump of spinal cord (Fig. 3C). Nerve

cell bodies were not found within the site of lesion at any

time during the postoperative period.

Nauta Stain

In the process of counting the number of fibers

in the three descending tracts following spinal cord

transaction, it was determined that the tectospinal and

thalamospinal tracts could not be effectively counted






44 -

separately because they were anatomically adjacent to one

another within the spinal cord. Consequently, they were

counted as one tract. There was little evidence of

degeneration 10 days postoperative to both the first

transaction and the subsequent retransection. This agreed

with lesion studies in the shark visual system utilizing the

Nauta technique where signs of degeneration following

lesions did not occur until approximately 20 days

postoperative (Ebbesson and Ramsey, 1968; Ebbesson and

Schroeder, 1971). Those sharks sacrificed at 20 days

following both transactions showed degeneration within the

three descending tracts and were thus used to make the counts.

Degenerating descending nerve fibers showed typical

irregular beading and droplet formations with concomitant

phagocytic activity typical of lower vertebrates. Ascending

tracts showed no retrograde degeneration.

Nerve fiber counts in the ventral cerebellospinal

tract and the combined tectospinal-thalamospinal tracts are

summarized in Table 2. Degenerating nerve fibers counted

20 days following the first transaction represent the normal

complement of axons in each of the respective tracts. Those

nerve fibers counted 20 days following the retransection at

90 days postoperative were the number of regenerated axons

that originated rostral to the lesion site. The number of





- 45 -


TABLE 2

SUMMARY OF DEGENERATING NERVE FIBER
COUNTS IN DESCENDING TRACTS


Tract Numbers of Fibers



Regenerated % Regenerated
Normal 90 Days 90 Days


Ventral Cerebellospinal 774 72 9.3

Tectospinal-Thalamospinal 1584 213 13.4






46 -

degenerating nerve fibers found within the combined

tectospinal-thalamospinal and ventral spinocerebellar tracts

following the second transaction were 13.4% and 9.3% respectively

of the number of degenerating nerve fibers found following the

first transaction.

Rasmussen Stain

The results of the synaptic terminal counts as

revealed by the Rasmussen stain on cell bodies and primary

dendrites of motor horn cells caudal to the lesion are

summarized in Table 3 and Fig. 5. The average number of

boutons represented was the combined data on both the left

and right sides of the spinal cord since there were no

statistically significant differences (P>.05) between

right and left counts. In addition, there was a high

correlation (r=.974, P<.01) between bouton counts on cell

bodies and primary dendrites.

The number of boutons on cell bodies were

statistically less than normal (P<.05) throughout the

postoperative period, including the retransected groups.

There was, however, a statistically significant increase

in boutons with time after surgery from 10 to 60 days

(P<.05), but no statistically significant difference occurred

between 60, 90 and the two retransected groups (P>.05).

Synaptic terminals dropped to 45% of normal at 10 days






- 47 -


TABLE 3

SUMMARY OF COMBINED LEFT-RIGHT BOUTON COUNTS
ON MOTOR HORN CELL BODIES
AND PRIMARY DENDRITES


Combined
Left-Right
Cell Body
Days Count
Postoperative (X+_SEM)

10 14.15+0.82

20 17.50+0.47

30 20.50+0.72

40 23.86+0.25

60 28.96+0.32

90 29.90+0.12

RETRANSECTION AT

10 29.90+0.49

20 29.78+0.49

Normal 31.40+0.48


of
Normal

45.06

54.61

65.28

75.98

92.22

95.22

90 DAYS

95.22

94.84


Combined
Left-Right
Count Per 10v
Primary Dendrite
(+ SEM)

5.30+0.49

5.09+ 0.98

5.92+0.42

6.58+0.17

7.30+0.38

7.78+0.37

POSTOPERATIVE

7.42+0.39

7.80+0.27

8.51+0.15


of
Normal

62.27

59.81

69.56

77.32

85.78

91.42



87.19

91.65























Figure 5 Results of bouton counts as revealed by the
Rasmussen stain. There was a high correlation
(r=.97, P<01) between counts on motor horn
cell bodies and primary dendrites.
































NORMAL




RALCELL BODY



CELL BODY


DENDRITE


RETRANS
o0


.10


z z
*8 -4 w
Z i
O
w 4
-O
*6 n
00



.4 -< Z
0
m *5
2 -
02
n -


- 49 -


35-



28-



21-



14-



7-


I I I I I I I I I I I
10 20 30 40 50 60 70 80 90 10 20
DAYS POSTOPERATIVE


Nc






50 -

postoperative, increased to 92% of normal by 60 days

postoperative and were 95.2% of normal at 90 days

postoperative. The retransected groups sacrificed at 10 and

20 days following retransection were 95.2% and 94.8% of

normal respectively.-

The number of boutons per 10 primary dendrite

following spinal cord transaction were significantly less

than normal from 10 to 60 days postoperative (P<.05) and

were significantly greater (P
(62.2% of normal at 10 days, 85.7% of normal at 60 days).

The bouton count was 91.4% of normal and statistically

indistinguishable from normal (P>.05) at 90 days

postoperative. Ten days following the retransection,

however, the number of boutons was 87.2% of normal and

statistically less than normal (P<~05). In contrast, there

was no statistically significant difference between normals

and 20 days following the retransection (P>.05).















BEHAVIOR

Daily Observations


Immediately upon recovery from anesthesia, all

operated sharks exhibited undulatory movements caudal to

the site of spinal cord transaction. This phenomenon

persisted throughout the postoperative period and was never

observed in normal animals. Although these undulatory

movements occurred caudal to the lesion, there were no

swimming movements caudal to the lesion when the animals

attempted to swim. Forward movement was accomplished by

"walking" along the bottom of the tank using the pectoral

fins or by jerking the body rostral to the lesion left and

right while dragging the caudal portion of the body. Turning

could only be done by walking movements of the pectoral fins.

The undulatory movements during the early

postoperative days (1-30 days) were not strong enough to

move the operated animals. From 30 days postoperative,

however, undulatory movements became strong enough to

propel the sharks forward. The animals did not appear

able to control these undulatory movements. As a result,

51 -






- 52 -


the undulatory movements proved detrimental to their swimming

ability. Stimulation caudal to the lesion site by gentle

prodding or by an inadvertent touch by another shark caused

an increase in undulatory strength which either flipped the

shark over on its back, using the snout as a pivotal point,

or pushed the startled animal into a wall despite its best

efforts to prevent this by using the pectoral fins to

"backpedal" away from the wall. The sharks remained

paralyzed caudal to the transaction for the duration of the

postoperative period with respect to normal attempts to swim.

Quantitative Tests

The results of the strength tests and timed swimming

trials are summarized in Table 4.

The strength.tests (Fig. 2) showed two types of

responses. The first response was elicited following

stimulation and was in the form of a single, sharp flexure

of the axial musculature. A second, consecutive stimulation

elicited the same type of response in the opposite direction

(Fig. 6B). This response following stimulation was present

in both normal and operated animals, although much reduced

in the operated animals. This response in operated animals

was probably due to passive conduction of movements caused

by muscle contractions rostral to the lesion. The second

type of response was the previously mentioned undulatory




- 53 -


TABLE 4

SUMMARY OF STRENGTH TEST RESULTS
FOLLOWING SPINAL CORD TRANSACTION


STRENGTH OF
RESPONSE TIMED
UNDULATORY FOLLOWING SWIMMING
STRENGTH STIMULATION % TRIALS %
DAYS IN KGMS IN KGMS OF IN SECONDS OF
POSTOPERATIVE (XSEM) (+MSEM) NORMAL (X+SEM) NORMAL


0.080.01
0.21+0.03
0.21 0.02
0.26+0.02
0.37+0.04
0.47 0.06
0.870.12
1.14 0.12
1.77+0.18
1.96 0.24
2.66 0.30
2.13 0.23
2.23 0.21
2.62 0.63
2.670.20
2.60 0.18
2.46+ 0.18


0.320.04
0.290.03
0.320.04
0.270.03
0.340.04
0.310.02
0.350.03
0.380.04
0.480.05
0.500.05
0.490.03
0.340.02
0.350.02
0.360.05
0.350.03
0.360.01
0.380.02


10.63
9.63
10.63
8.97
11.29
10.29
11.62
12.63
15.94
16.61
16.27
11.29
11.62
11.96
11.62
11.96
12.63


9.300.30
8.860.27
8.73 0.44
8.89 0.38
9.840.47
8.700.44
8.060.37
7.850.30
7.840.28
8.430.36
8.440.20
8.730.31
8.170.38
9.170.55
8.710.32
9.080.19
9.030.09


RETRANSECTION AT 90 DAYS POSTOPERATIVE


2.28+ 0.16
2.320.19
2.44+ 0.18


0.310.01
0.310.02
0.32+0.01


10.29 10.450.17
10.29 9.750.32
10.63 9.660.26


N/A 3.010.32


16.12
16.70
16.95
16.64
15.04
17.01
18.36
18.85
18.87
17.55
17.53
16.95
18.11
16.13
16.99
16.29
16.38


14.16
15.17
15.32


Normal


1.48 0.22























Figure 6 A. Polygraph tracing showing undulatory movements
(0.25-0.5 cycles/sec) caudal to transaction
site. The top trace indicates when stimulation
occurred. Paper speed = 3mm/sec.

B. Polygraph tracing showing response of axial
musculature following stimulation of the barbels
of a normal animal. The top trace indicates
when stimulation occurred. Paper speed = 3mm/se





- 55 -


A










B





56 -

movements caudal to the site of lesion. These undulatory

movements occurred only in operated animals and were an

almost constant phenomenon requiring no stimulation

(Fig. 6A).

The response following stimulation was significantly

reduced from normal throughout the postoperative period

(P<.001) following the first transaction and there was no

trend during this postoperative period for any return of

strength caudal to the lesion. There was also no statistically

significant difference (P>.05) between the response following

stimulation at 90 days postoperative and the retransected

groups (Fig. 7A).

Undulatory movements (0.25-0.5 cycles/sec) were

weak during the early postoperative period, but there was a

statistically significant increase in the strength of

undulatory movements with increased time from 20 to 60 days

postoperative (P<.05). From 60 days postoperative to 20

days following the retransection, there were no significant

differences between postoperative groups (P>.05). In addition,

the strength attained by undulatory movements at 60 days

postoperative was statistically indistinguishable (P>.05)

from the strength of response following stimulation of

normal animals (Fig. 7A).

A significant increase was observed in the time
























Figure 7 A. Results of strength tests following spinal
cord transaction.

B. Results of timed swimming trials following
spinal cord transaction.







- 58 -


3.0' NORMAL
3.0 3.0

Z w
O m

* E ANS


-n 1.8 6- .8J .6

UNDUL
- 1.2 5 1. .1.2 S _









DAYS POSTOPERATIVE A



11--
i0 2









S10











0 a-




S PRETRANS
0 4-



[ 2- NORMAL


I i I I I I 2 I
10 20 30 40 50 60 70 80 90 10 20
DAYS POSTOPERATIVE






59 -

required to swim seven feet following spinal cord transaction

(P<.001). Normal animals required an average of 1.48 seconds,

whereas the mean value for transected animals was 8.72

seconds. There was no trend during the postoperative period

for improvement in swimming times (Fig. 7B) and there were

no significant differences between times following the

first and second spinal cord transactions (L>.05).















DISCUSSION


These results indicate that following spinal cord

transaction, the nurse shark is capable of limited anatomical

regeneration of descending tracts across the site of lesion

(9.3-13.4%) to an area six spinal segments caudal to the

lesion. The time required for nerve fibers to regenerate

across the lesion site was between 40 and 60 days. This was

considerably slower and more incomplete than regeneration in

the teleost spinal cord. Teleosts have shown anatomical

regeneration and return of normal swimming function from

four days postoperative in guppies (Hooker, 1930, 1932)

to approximately 35 days postoperative in goldfish (Bernstein,

1964) although descending fiber tracts in the goldfish

regenerate only 35-49% of the original complement of fibers

within a given tract (Bernstein and Gelderd, 1970).

Limited regeneration did occur in the shark but there was

no return of strength in the axial musculature caudal to

the spinal cord transaction following stimulation rostral

to the lesion. The operated sharks were paralyzed caudal

to the lesion in normal attempts to swim. This lack of

60 -






61 -

strength caudal to the lesion was also reflected in the

timed swimming trials as there was no trend during the

postoperative period for improvement in swimming times.

In fact, the small amount of anatomical regeneration seen

six spinal segments caudal to the lesion at 90 days

postoperative had little or no effect on strength or

swimming speed as a subsequent retransection at 90 days

postoperative caused no change in these performance parameters

in the ensuing postoperative period.

As was stated previously, only a small number of

nerve fibers regenerated rostro-caudal across the site of

lesion to the area six spinal segments caudal to the lesion

and they did not traverse the lesion site until 40 to 60

days following transaction. Despite this, the number of

synapses on motor horn cell bodies and primary dendrites

caudal to the lesion showed an increase beginning at 20 days

postoperative through 60 days postoperative and were highly

correlated (r=.974, P<.01).

If synaptic return occurs before the return of

long tract input, what is the source of the increase in the

number of synapses on motor horn cells caudal to the lesion?

The high correlation between primary dendrite and cell body

synaptic counts strongly indicates that both phenomena have

the same origin. The origin for the nerve fibers which






62 -

replaced lost synaptic contacts must have been caudal to

the site of lesion in the form of local, segmental sprouting.

This hypothesis is further supported by data following the

retransection at 90 days postoperative. There was no

significant change in the number of boutons on motor horn

cell bodies following retransection and a small but statistically

significant (P<.05) drop (91.4% to 87.6% of normal) in boutons

on motor horn cell primary dendrites. This slight drop in

boutons on primary dendrites following retransection was

probably due to degeneration of the small number of regenerated

long tract nerve fibers which synapsed on motor horn cell

dendrites.

If there is no return of swimming prowess and no

return of axial musculature strength following stimulation,

then the question arises as to the functional significance

of this sprouting phenomenon and the return of synaptic

contacts on motor horn cells caudal to the lesion site.

Perhaps the answer to this question lies in the unique

appearance in sharks of undulatory movements caudal to the

lesion following spinal cord transaction. Before this

relationship is discussed, however, the anatomical basis

for the undulatory movements will be elaborated.

Unlike other shark studies which claim coordinated

undulatory movements propagated rostro-caudally over the






63 -

site of lesion (Ten Cate and Ten Cate-Kazejawa, 1933;

Gray, 1936; Gray and Sand, 1936a, 1936b), undulatory movements

in this experiment were observed only caudal to the lesion

site and were independent of body movements rostral to the

lesion. In fact, during the strength tests, the response

following stimulation was often superimposed upon undulatory

movements without affecting the speed or strength of the

undulatory movements.

There are at least two hypotheses relating to the

anatomical basis of undulatory movements in spinal sharks.

Ten Cate and Ten Cate-Kazejawa (1933) claim that the

undulatory movements are propagated over the lesion site

by tensile stimuli applied to posterior musculature when an

active contraction occurs in the head region, implying the

activity of a chain of peripherally controlled reflexes.

Gray (1936) and Gray and Sand (1936a, 1936b) showed that

coordinated responses did not occur if two regions of the

body of a dogfish were isolated from one another by a

second spinal cord transaction. Each isolated section of

the body exhibited a spontaneous, independent undulatory

activity. Gray and Sand attributed this to an inherent

undulatory discharge rhythm within the spinal cord.

Lissman (1946a, 1946b) has shown that undulatory movements

caudal to a spinal cord transaction in dogfish can only be






64 -

abolished by a complete bilateral rhyzotomy caudal to the

lesion. The hypothesis of Ten Cate and Ten Cate-Kazejawa

(1933) must be rejected in the present experiment because

undulatory movements persisted without movement rostral to

the lesion. In fact, the undulatory movements were most

prevalent when the shark was at rest on the bottom and were

absent in normal attempts to swim. The two theories need

not be mutually exclusive, however. Lissmann's studies

showed that dorsal root input is mandatory for the maintenance

of undulatory movements. Thus it appears from the present

data that the spinal cord of the shark has an inherent

undulatory discharge pattern modulated by local sensory

input and input from brain centers. If the major effect

of the brain on this undulatory discharge pattern is

inhibitory, transecting the spinal cord will release the

caudal section of spinal cord from these inhibitory influences,

thus allowing the inherent discharge pattern to be exhibited

in the form of undulatory movements caudal to the lesion site.

If the synaptic sites left vacant by spinal cord transaction

are replaced with excitatory synapses from dorsal root

fibers or indigenous spinal tracts by way of sprouting, an

increase in the discharge pattern should occur with a

resultant increase in the strength of undulatory movements.

Undulatory movements in the present experiment began






65 -

immediately upon recovery from anesthesia and increased

in strength up to 60 days postoperative after which they

leveled off. If the postoperative increase in the strength

of undulatory movements is compared to the postoperative

increase in synaptic complement on motor horrf cells caudal

to the lesion and the resultant data plotted on a graph

(Fig. 8A, 8B), the curves of the two postoperative phenomena

are highly correlated. Comparison between boutons on cell

bodies and undulatory movements has a correlation coefficient

of r=.930 (L<.01), and comparison between boutons on motor

horn cells primary dendrites and undulatory movements results

in a correlation coefficient of r=.91 (P<.01).

It is therefore highly probable from these data

that the synaptic return on motor horn cells by way of

local sprouting is responsible for the increase in the

strength of undulatory activity caudal to the site of lesion.

To further support this hypothesis, retransection at 90 days

had a minimal effect on both synaptic complement and

undulatory movements (Fig. 8A, 8B).
























Figure 8 A. Comparison between synapse count on motor
horn cell bodies caudal to the lesion and
undulatory strength (r=.93, P<.01).

B. Comparison between synapse count on motor
horn cell dendrites caudal to the lesion
and undulatory strength (r=.91, P<.01).








- 67 -


BOUTONS



UNDUL
W-m--nlk


A jwr RETRANS
0 *

I I I I I I I I I I
10 20 30 40 50 60 70 80 90 10 20
DAYS POSTOPERATIVE


DOUTONS



UNDUL
*----*


1 *^*- RETRANS

I I I I I I I I I I
10 20 30 40 50 60 70 80 90 10 20
DAYS POSTOPERATIVE


-3.0
-4
m
Z
w2.4 g
x
"4



-1.8 8

m


C
-1.2

zA
0
.0.6 -

0
a


-4
a

-2.4 3
-t
Oz
-1.8 e

Z u
U 0
-1.2 O
c
0
C
-0.6

a














CONCLUSION


Regeneration in the shark spinal cord following

spinal cord transaction appears to lie somewhere between

the abortive regeneration usually seen in mammals and the

vigorous regeneration and return of function typical of

teleosts. Although anatomical regeneration of nerve

fibers across the site of lesion does occur, the functional

ramifications are negligible and for all practical purposes

the shark remains paralyzed caudal to the lesion when

attempting to swim. The poor anatomical regeneration and

lack of functional return in sharks is surprising. A

general rule of thumb which is well documented is that

the lower on the phylogenetic scale, the more vigorous

and complete is the central nervous system regenerative

process. The reasoning for this is that more primitive animals

reportedly possess more undifferentiated, pleuri-potential

cells capable of differentiation into neural elements. The

elasmobranchs occupy the third rung up on the vertebrate

phylogenetic ladder immediately below the teleosts, yet show


- 68 -






69 -

a postoperative recovery following spinal cord transaction

more akin to mammals than fish.

Perhaps the most significant and interesting

result of this experiment is the strong indication of a

functional correlation between the return of synapses on

motor horn cells caudal to the lesion and the increase in

undulatory strength during the postoperative period.

Although functional correlates have been suggested for the

phenomenon of sprouting in the mammalian central nervous

system (McCouch et al., 1955; Schneider, 1970) the evidence

presented in this experiment is perhaps the most conclusive

to date of a functional correlate to sprouting in the

vertebrate central nervous system.















LITERATURE CITED


Adams, J. H., P. M. Daniel and M. M. L. Prichard 1968
Regrowth of nerve fibers in the neurohypophysis:
Regeneration of a tract of the central nervous system.
J. Physiol. (London), 198:4P-5P.

Adams, J. H., P. M. Daniel and M. M. L. Prichard 1969
Degeneration and regeneration of hypothalamic nerve
fibers in the neurohypophysis after pituitary stalk
section in the ferret. J. Comp. Neurol., 135:121-144.

Adams, J. H., P. M. Daniel and M. M. L. Prichard 1971
Changes in the hypothalamus associated with regeneration
of the hypothalamo-neurohypophysial tract after
pituitary stalk section in the ferret. J. Comp. Neurol.,
142:109-124.

Aronson, L. R. 1963 The central nervous system of sharks
and bony fishes with special reference to sensory and
integrative mechanisms. In Sharks and Survival (Ed.
Perry W. Gilbert) pp. 165-241, D. C. Heath and Co.,
Boston, Massachusetts.

Attardi, D. G. and R. Sperry 1963 Preferential selection
of central pathways by regenerating optic fibers.
Exp. Neurol., 7:46-64.

Bernstein, J. J. 1964 Relation of spinal cord regeneration
to age in adult goldfish. Exp. Neurol., 9:161-174.

Bernstein, J. J. 1970 Anatomy and Physiology of the central
nervous system. In Fish Physiology Vol. IV (Ed. W. S.
Hoar and D. J. Randall) pp. 1-90, Academic Press,
New York, London.

Bernstein, J. J. and M. E. Bernstein 1967 Affect of the
glial ependymal scar and teflon arrest on the
regenerative capacity of the goldfish spinal cord.
Exp. Neurol., 19:25-32.






- 71 -


Bernstein, J. J. and M. E. Bernstein 1968 Contact inhibition:
A mechanism of abortive regeneration in the goldfish
spinal cord. Anat. Rec. 160:315-316.

Bernstein, J. J. and M. E. Bernstein 1971 Axonal regeneration
and formation of synapses proximal to the site of lesion
following hemisection of the rat spinal cord. Exp.
Neurol., 30:336-351.

Bernstein, M. E. and J. J. Bernstein 1972 Regeneration in
the spinal cord of the monkey. (In press).

Bernstein, J. J. and J. B. Gelderd 1970 Regenerative
capacity of long spinal tracts in the goldfish. Brain
Res., 19:21-26.

Bernstein, J. J. and J. B. Gelderd 1972 Synaptic complement
formation following regeneration of the goldfish spinal
cord. (In manuscript).

Bjorklund, A. R. Katzman, U. Stenevi and K. A. West 1971
Development and growth of axonal sprouts from noradrenaline
and 5-hydroxytryptamine neurones in the cat spinal cord.
Brain Res., 31:21-33.

Bjorklund, A. and U. Stenevi 1971 Growth of central
catecholamine neurones into smooth muscle grafts in
the rat mesencephalon. Brain Res., 31:1-20.

Brown, J. 0. and G. P. McCouch 1947 Abortive regeneration
of the transected spinal cord. J. Comp. Neurol., 87:131-137.

Cattaneo, D. 1923 I fenomeni degenerativi nelle vie visive
in seguito a lesioni del nervo ottico. Riv. Pat. Nerv.
28:61-118.

Clemente, C. D. 1955 Structural regeneration in the mammalian
central nervous system and the role of neuroglia and
connective tissue. In Regeneration in the Central Nervous
System (Ed. William F. Windle) pp. 147-161, Charles C.
Thomas, Springfield, Illinois.

Clemente, C. D. 1964 Regeneration in the vertebrate central
nervous system. Intern. Rev. Neurobiol., 6:257-301.

Drummond, C. D., Jr. 1954 The influence of piromen on the
regeneration of the spinal cord in adult Triturus
viridescens. Undergrad. Hon. Thesis, Brown Univ.,
ASt'v 0 a''Q~-; r>. i.






- 72 -


Ebbesson, Sven O. E. 1972 New insights into the organization
of the shark brain. Comp. Biochem. and Physiol. 42:121-130.

Ebbesson, Sven O. E. and John S. Ramsey 1968 The optic
tracts of two species of sharks (Galeocerdo cuvier and
Ginglymostoma cirratum). Brain Res. 8:36-53.

Ebbesson, Sven O. E. and Dolores M. Schroeder 1971
Connections of the nurse shark's telencephalon. Science
173:254-256.

Freeman, L. W. 1955 Functional recovery in spinal rats.
In Regeneration in the Central Nervous System (Ed.
William F. Windle) pp. 195-207, Charles C. Thomas,
Springfield, Illinois.

Fridberg, G., R. S. Nishioka and W. R. Fleming 1966
Regeneration of the caudal neurosecretory system in
the cichlid teleost Tilapia mossambica. J. Exp.
Zool. 162:311-336.

Gaze, R. M. 1970 The Formation of Nerve Connections.
Academic Press, London and New York.

Gaze, R. M. and M. Jacobson 1963 The path from the retina
to the ipsilateral optic tectum of the frog. J.
Physiol. 165:73-74.

Gaze, R. M. and M. J. Keating 1969 The depth distribution
of visual units in the tectum of the frog following
regeneration of the optic nerve. J. Physiol. 200:128-129.

Gaze, R. M. and M. J. Keating 1970a Further studies on the
restoration of the contralateral retinotectal projection
following regeneration of the optic nerve in the frog.
Brain Res., (In Press).

Gaze, R. M. and M. J. Keating 1970b Regenerated visual
units in the frog. Brain Res., (In Press).

Goldstein, L. 1967 Urea Biosynthesis in Elasmobranchs.
In Sharks, Skates and Rays (Ed. Perry W. Gilbert,
Robert F. Mathewson and David P. Rall) John Hopkins
Press, Baltimore, Maryland.

Goldstein, L., W. W. Oppelt and T. H. Maren 1968 Osmotic
regulation and urea metabolism in the lemon shark
Negaprion brevirostris. Am. J. Physiol. 215:1493,1497.






- 73 -


Goodman, D. C. and J. A. Horel 1966 Sprouting of optic
tract projections in the brain stem of the rat.
J. Comp. Neurol., 127:71-78.

Gray, J. 1936 Studies in animal locomotion. IV. The
neuromuscular mechanism of swimming in the eel.
J. Exp. Biol., 13:170.

Gray, J. and A. Sand 1936a The locomotory rhythm of the
dogfish (Scyllium canicula). J. Exp. Biol. 13:200-209.

Gray, J. and A. Sand 1936b Spinal reflexes of the dogfish
(Scyllium canicula). J. Exp. Biol. 13:210-217.

Guth, L. and W. F. Windle 1970 The enigma of central nervous
regeneration. Exp. Neurol. 28:1-43, Supp. 5.

Hamburger, V. 1955 Regeneration in the central nervous
system of reptiles and of birds. In Regeneration in
the Central Nervous System (Ed. William F. Windle)
pp. 47-53, Charles C. Thomas, Springfield, Illinois.

Harvey, J. E. and H. H. Srebnik 1967 Locomotor activity
and axon regeneration following spinal cord compression
in rats treated with L-thyroxine. J. Neuropath. Exp.
Neurology 26:661-668.

Healey, E. G. 1957 The Nervous System. In The Physiology
Of Fish Vol. 2 (Ed. Margaret E. Brown) pp. 1-119,
Academic Press Inc., New York.

Healey, E. G. 1962 Experimental evidence for regeneration
following spinal section in the minnow (Phoxinus phoxinus).
Nature 194:395-396.

Hibbard, E. 1963 Regeneration of the severed spinal cord
of chordate larvae of Petromyzon marinus. Exp. Neurol.,
7:175-185.

Hooker, D. 1925 Studies on regeneration in the spinal cord.
III. Reestablishment of anatomical and physiological
continuity after transaction in frog tadpoles. J.
Comp. Neurol. 38:315-347.

Hooker, D. 1930 Physiological reactions of goldfish with
severed spinal cord. Proc. Soc. Exp. Biol. Med.,
28:89-90.






- 74 -


Hooker, D. 1932 Spinal cord regeneration in the young
rainbow fish Lebistes Reticulatus. J. Comp. Neurol.,
56:277-295.

Order, T. J. 1971 Retention by fish optic nerve fibers
regenerating to new terminal sites in the tectum of
"chemospecific" affinity for their original sites.
J. Physiol. (London) 216:53-55P.

Jacobson, M. and R. M. Gaze 1965 Selection of appropriate
tectal connections by regenerating optic nerve fibers
in adult goldfish. Exp. Neurol., 13:418-430.

Kappers, Ariens C. U. 1906 The structure of the teleostian
and selachian Brain. J. Comp. Neurol. and Psychology.
16:1-110.

Kappers, Ariens C. U., G. C. Huber and E. C. Crosby 1936
The Comparative Anatomy of the Nervous System of
Vertebrates Including Man. Vol. 1, The Macmillan Co.,
New York.

Katzman, R., A. Bjorklund, Ch. Owman, U. Stenevi and K. A.
West 1971 Evidence for regenerative axon sprouting
of central catecholamine neurons in the rat mesencephalon
following electrolytic lesions. Brain Res., 25:579-596.

Keil, J. H. 1940 Functional spinal cord regeneration in
adult Rainbow fish. Soc. Exp. Biol. Med., 43:175-177.

Kirsche, W. 1950 Die regenerativen Vorgange am Ruckenmark
eruachsever Teleostier nach operativer Kontinuitat.
Strennung. Z. Mikroskop. Anat. Forsch. 56:190-265.

Kirsche, W. 1956 Experimentelle Untersuchunger uber die
regeneration des durchtrennten Ruckenmarkes von
Amblystoma mexacanum. Z. Mikroskop. Anat. Forsch.
65:512-586.

Kirsche, W. 1965 Regenerative Vorgange im Gehirn und
Ruckenmark. Ergeb. Anat. Entiv-gesch. 38:143-194.

Koppanyi, T. 1955 Regeneration in the central nervous
system of fishes. In Regeneration in the Central
Nervous System (Ed. William F. Windle) pp.3-19,
Charles C. Thomas, Springfield, Illinois.





- 75 -


Koppanyi, T. and P. Weiss 1922 Funktionelle regeneration
des Ruckenmarkes bei anamniern. Anz. Akad. Wiss. Wien.
Math. Naturre Kl. 59:206-219.

Levi-Montalcini, R. and B. Brooks 1960 Excessive growth
of the sympathetic ganglis evoked by a protein isolated
from the mouse salivary gland. Proc. Nat. Acad. Sci.
(Wash.) 46:373-384.

Lissmann, H. W. 1946a The neurological basis of the
locomotory rhythm in the spinal dogfish (Scyllium
canicula, Acanthius vulgaris) I. Reflex Behavior.
J. Exp. Biol. 23:143-161.

Lissmann, H. W. 1946b The neurological basis of the
locomotory rhythm in the spinal dogfish (Scyllium
canicula, Acanthius vulgaris) II. The effect of
deafferentation. J. Exp. Biol. 23:162-176.

Littrell, J. L. 1955 Apparent functional restitution in
piromen treated spinal cats In Regeneration in the
Central Nervous System (Ed. William F. Windle)
pp. 219-228, Charles C. Thomas, Springfield, Illinois.

Littrell, J. L., D. Bunnell, W. F. Agnew, J. O. Smart and
W. F. Windle 1953 Effects of a bacterial pyrogen on
hind-limb function in spinal cats. Anat. Rec., 115:430-436.

Liu, C. N. and W. W. Chambers 1958 Intraspinal sprouting of
dorsal root axons. Arch. Neurol. 79:46-61.

Lund, R. D. and J. S. Lund 1971 Synaptic adjustment after
deafferentation of the superior colliculus of the rat.
Science, 171:804-807.

Maron, K. 1959 Regeneration capacity of the spinal cord
in Lampetra fluviatilis larvae. Folia Biol., 7:179-189.

McCouch, G. P., G. M. Austin and C. Y. Liu 1955 Sprouting
of new terminals as a cause of spasticity. Am. J.
Physiol. 183:642.

Moore, R. Y., A. Bjorklund and U. Stenevi 1971 Plastic
changes in the adrenergic innervation of the rat
septal area in response to denervation. Brain Res.
(In Press).





- 76 -


Niazi, I. A. 1963 The histology of tail regeneration in
the ammocoetes. Can. J. Zool., 41:125-145.

Nicholas, J. S. and D. Hooker 1928 Progressive cord
degeneration and collateral transmission of spinal
impulses following section of the cord in albino
rat fetuses. Anat. Rec. 38:24-32.

Nieuwenhuys, R. 1964 Comparative anatomy of the spinal
cord. Prog. Brain Res. 11:1-57.

Pearcy, J. F. and T. Koppanyi 1924 A further note on
regeneration of the cut spinal cord in fish. Proc.
Soc. Exp. Biol. Med., 22:17-19.

Piatt, J. 1955a Regeneration in the central nervous
system of amphibia. In Regeneration in the Central
Nervous System (Ed. William F. Windle) pp. 20-46,
Charles C. Thomas, Springfield, Illinois.

Piatt, J. 1955b Regeneration of the spinal cord in the
salamander. J. Exp. Zool., 129:177-207.

Piatt, J. and M. Piatt 1958 Transection of the spinal
cord in the adult frog. Anat. Rec. 131:81-95.

Raisman, G. 1966 The connexions of the septum. Brain
89:317-348.

Raisman, G. 1969a A comparison of the mode of termination
of the hippocampal and hypothalamic afferents to the
septal nuclei as revealed by the electron microscopy
of degeneration. Exp. Brain Res. 7:317-343.

Raisman, G. 1969b Neuronal plasticity of the septal nuclei
of the adult rat. Brain Res. 14:25-48.

Ramon y Cajal, S. 1928 Degeneration and Regeneration of the
Nervous System. 1:47-51.

Ramon y Cajal, S. 1960 Studies on Vertebrate Neurogenesis,
L. Guth, Translator, Charles C. Thomas, Springfield,
Illinois.






- 77 -


Rasmussen, G. L. 1957 Selective silver impregnation of
synaptic endings In New Research Techniques of
Neuroanatomy, (Ed. W. F. Windle) pp. 27-39, Charles
C. Thomas, Springfield, Illinois.

Rasmussen, L. E. 1971 Organ distribution of exogenous
14C-urea in elasmobranchs with special regard to the
nervous system. Comp. Biochem. Physiol. 40A:145-154.

Rasmussen. L. E. and R. A. Rasmussen 1967 Comparative
protein and enzyme profiles of the cerebrospinal fluid,
extradural fluid, nervous tissue and sera of
elasmobranchs In Sharks, Skates and Rays (Ed. Perry W.
Gilbert, Robert F. Mathewson and David P. Rall)
pp. 361-380, John Hopkins Press, Baltimore, Maryland.

Schneider, G. E. 1970 Mechanisms of functional recovery
following lesions of visual cortex and superior
colliculus in neonate and adult hamsters. Brain,
Behavior and Evol. 3:295-323.

Scott, D., Jr., E. Gutmann and P. Horsky 1966 Regeneration
in spinal neurons: Proteosynthesis following nerve
growth factor administration. Science 152:787-788.

Scott, D. and C. N. Liu 1964 Factors promoting regeneration
of spinal neurons: Positive influence of nerve growth
factor. Prog. in Brain Res. 13:127-150.

Segaar, J. 1965 Behavioral aspects of degeneration and
regeneration in fish brain: A comparison with higher
vertebrates. Prog. Brain Res. 14:143-231.

Sharma, S. C. 1972 Reformation of retinotectal projections
after various tectal ablations in goldfish. Exp.
Neurol. 34:171-182.

Smith, H. 1929 The composition of the body fluids in
elasmobranchs. J. Biol. Chem. 81:407-419.

Sperry, R. W. 1944 Optic nerve regeneration with return
of vision in anurans. J. Neurophysiol. 7:57-70.

Sugar, O. and R. W. Gerard 1940 Spinal cord regeneration
in the rat. J. Neurophysiol. 3:1-19.






- 78 -


Ten Cate, J. and Ten Cate-Kazejawa 1933 La coordination des
movements locomoteurs apres la section transversale de
la moelle epiniere chez les requins. Arch. Neerl.
Physiol. 18:15-23.

Von Lenhossek, M. 1892 Beobachtungen an den spinalganglien
und dem Ruchenmark von Pristiurusembryonen. Anat.
Anz. Bd. 7:519.

Von Lenhossek, M. 1895 Der fienere bau des nervensystems
im lechte neuester Forschungen. Aufl. Fischer, Berlin.

Windle, W. F. 1955 Comments on regeneration in the human
central nervous system. In Regeneration in the Central
Nervous System (Ed. William F. Windle) pp. 265-272,
Charles C. Thomas, Springfield, Illinois.

Windle, W. F. 1956 Regeneration of axons in the vertebrate
central nervous system. Physiol. Rev. 36:427-440.

Windle, W. F. and W. W. Chambers 1950a Regeneration in the
spinal cord of the cat and dog. J. Comp. Neurol.
93:241-257.

Windle, W. F. and W. W. Chambers 1950b Spinal cord
regeneration associated with a cellular reaction induced
by administration of a purified bacterial pyrogen.
Abst. V. Internat. Anat. Comg., Oxford, p. 196.

Windle, W. F. and W. W. Chambers 1951 Regeneration in the
spinal cord of the cat and dog. Arch. Neurol. Psychiat.,
Chi., 65:261-262.

Windle, W. F., C. D. Clemente and W. W. Chambers 1952
Inhibition of formation of a glial barrier as a means
of permitting peripheral nerve to grow into the brain.
J. Comp. Neurol. 96:359-370.

Yoon, M. 1971 Reorganization of retinotectal projection
following surgical operations on the optic tectum in
goldfish. Exp. Neurol. 33:395-411.















BIOGRAPHICAL SKETCH

John Bruce.Gelderd was born during a German

bombing raid in Wasquehal, France on September 21, 1939.

He attended grade school in Broadview, Illinois, and Clearwater,

Florida, and graduated from Clearwater High School in 1957.

After a 4 year stint in the U. S. Air Force, he attended

St. Petersburg Jr. College from September, 1962, to May,

1964. He entered the University of Florida in September,

1964 and received a bachelor's degree in zoology in 1967.

He began his graduate studies toward the Doctor of

Philosophy degree at the University of Florida in September,

1968.

He has accepted a faculty position in the

Department of Anatomy at Louisiana State University in

New Orleans, Louisiana.


- 79 -








I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.



(J J. Bernstein, Chairman
Associate Professor of Neuroscience

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
/


F. A. King, Chairman of
Department of Neuroscience


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality
as a dissertation for the degree of Doctor of Philosophy.



C.j/. Vierck,
A ciate Pr essor of Neuroscience


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.



John B. Munson
iAssistant Professor, Neuroscience
1/







This dissertation was submitted to the Department of
Neuroscience in the College of Medicine and to the Graduate
Council, and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.


August, 1972


Dean, CoLlege of


Dean, Graduate School
































UNIVERSITY OF FLORIDA

3I I126I 2 08554 3782 III ill I
3 1262 08554 3782