Axonal interactions between fetal spinal cord transplants and the adult rat spinal cord


Material Information

Axonal interactions between fetal spinal cord transplants and the adult rat spinal cord
Physical Description:
viii, 218 leaves : ill. ; 29 cm.
Jakeman, Lyn Burrell, 1961-
Publication Date:


Subjects / Keywords:
Research   ( mesh )
Neuronal Plasticity   ( mesh )
Spinal Cord -- transplantation   ( mesh )
Axons -- growth & development   ( mesh )
Rats   ( mesh )
Fetal Tissue Transplantation   ( mesh )
Department of Neuroscience thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Neuroscience -- UF   ( mesh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida, 1990.
Bibliography: leaves 194-217.
Statement of Responsibility:
by Lyn Burrell Jakeman.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001483999
oclc - 22503288
notis - AGZ6058
System ID:

Full Text








This undertaking is one which could not have been

completed if it were not for the help of several people.

First and foremost has been my mentor, Dr. Paul Reier, for

whom I have the utmost respect and admiration for his advice

and role as teacher and scientist. He contributed an

invaluable amount of time and patience to my training and

taught me the importance of maintaining a balance between

persistence and flexibility; a lesson that is vital in

research, writing, and communication in science. Additional

appreciation is extended to each member of my committee --

Drs. Barbara Bregman, John Munson, Roger Reep, Lou Ritz, Don

Stehouwer, and Chuck Vierck -- for continued support and

constructive criticisms regarding my work.

The daily progress was made more pleasant with the superb

technical and organizational assistance of Barbara O'Steen,

Minnie Smith, and Regina Reier, who kept track of my loose

pieces of paper and saved me months of effort. Additional

help was provided by the secretarial and support personnel in

the Departments of Neuroscience and Neurosurgery. Gratitude

is also extended to fellow graduate students, especially

Denise and Greg, who taught me to believe in myself when the

game seemed lost. Oversight of animal care and use was kept

by Dr. Dan Theele, D.V.M.. Finally, instruction in the

concepts of morphometric image analysis was made available by

Mr. Dan Williams.

An incredible amount of emotional support has been

extended over the past six years by family and friends;

especially my sister Barb, my parents, and my in-laws.

However, the greatest appreciation is extended to my husband,

David T. Lee, who provided both support and encouragement,

and also helped with editorial suggestions, technical

assistance, and scientific criticism.

The last personal acknowledgements go to Bob Yant and

Jim Sutherland, who provided needed reminders of the

importance of developing a progressive outlook with regard to

spinal cord research.

Financial support for equipment, supplies, and the much

needed student assistantship was provided by NIH grants 22316

and NS72300 to P.J. Reier, The Mark F. Overstreet Fund for

Spinal Cord Regeneration Research, The Center for

Neurobiological Sciences (NIMH grant MH15737), and the

Department of Neuroscience. Additional travel support was

provided by the American Paralysis Association.



ACKNOWLEDGEMENTS.......................................... ii


ABSTRACT ................................................vii


1 INTRODUCTION AND BACKGROUND ....................... 1

Spinal Cord Injury................................. 1
Treatments to Minimize Functional Loss............ 2
Promotion of Axonal Regeneration or
Sprouting ........................................ 4
Fetal Neural Transplants and Spinal
Cord Repair...................................... 8
Development of a Neural Relay Across a
Spinal Injury Site............................... 11
Experimental Goals ............................... 12

2 GENERAL METHODS .................................. 14

Experimental Animals.............................. 14
Preparation of Lesion Cavities ................... 15
Preparation of Donor Tissue ...................... 16
Transplantation Procedure......................... 16
Post-Operative Care............................... 17


Introduction ..................................... 18
Materials and Methods.............................20
Results .......................................... 23
Discussion .......................................43



Introduction ..................................... 55
Materials and Methods ............................ 57
Results........................................... 67
Discussion ...................................... 116

TRANSPLANTS ................................... 129

Introduction .................................... 129
Materials and Methods ........................... 132
Results.......................................... 139
Discussion ...................................... 168

6 SUMMARY AND CONCLUSIONS ......................... 180

Construction of a Relay Across a FSC Graft...... 180
Specificity Issues .............................. 182
Possible Role of FSC Grafts in Segmental and
Long-Tract Functions .......................... 187
Future Directions ............................... 191
Conclusions ..................................... 192

REFERENCES ............................................. 194

BIOGRAPHICAL SKETCH .................................... 218





FG -

FI -






NT -

Ox -




TH -


5-HT -


calcitonin gene-related peptide

central nervous system

corticospinal tract



Fusion Index

fetal spinal cord

gamma-aminobutyric acid

glial fibrillary acidic protein

horseradish peroxidase

myelin basic protein



peroxidase anti-peroxidase

Phaseolus vulqaris leucoagglutinin

peripheral nervous system

tyrosine hydroxylase

wheat germ agglutinin, conjugated to HRP

serotonin (5-hydroxy-tryptamine)

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



Lyn Burrell Jakeman

May, 1990

Chairman: Dr. Paul J. Reier
Major Department: Neuroscience

One approach to spinal cord repair involves the

transplantation of homotopic fetal neural tissue into the

site of a spinal lesion. As a strategy with potential toward

eventual functional recovery, this approach includes two

primary objectives. The first goal is to replace intrinsic

spinal cord elements at the lesion site. The second is to

provide a means for reconstructing functional continuity

between the separated rostral and caudal ends of the spinal

cord. Accordingly, the following series of experiments

establish an anatomical setting for functional repair.

To evaluate the potential for replacement of intrinsic

spinal cord tissue, intraspinal transplants of fetal spinal

cord (FSC) tissue were examined using conventional light and

electron microscopic techniques and immunocytochemical

staining. The normal substantial gelatinosa was compared with


distinct myelin-free regions of the grafts and the two were

found to contain similar cytological characteristics and

similar patterns of peptide staining.

Axonal projections between (FSC) grafts and the host

spinal cord were identified using a variety of neuro-

anatomical tracing and immunocytochemical techniques. The

presence of host fiber growth into the grafts, an extensive

pattern of intrinsic graft projections, and efferent growth

of axons into the host were consistent with the hypothesis

that fetal transplants may establish a neural relay across a

spinal cord lesion site.

Finally, interactions between the long myelinated fiber

tracts of the spinal cord and FSC grafts were examined using

the corticospinal tract (CST) as a model system. Injured CST

axons were observed in direct apposition to FSC tissue at the

host-graft interface, and CST axons were also seen extending

into the transplants.

Together, the results indicate that FSC transplants can

be used to restore anatomical continuity through 1) the

differentiation of intrinsic spinal cord regions at a lesion

site and 2) the development of axonal interactions between

host and graft tissues in the adult rat spinal cord. These

observations provide a basis for future studies to examine

the functional integration of FSC transplants, as well as a

model for investigating the biology of axonal elongation.




Spinal Cord Injury

Traumatic injury to the spinal cord begins a sequence of

events that result in the degeneration of spinal cord tissue

and subsequent loss of sensory and motor function. In order

to understand the clinical picture, the underlying

pathophysiology has been studied extensively in experimental

models of spinal cord injury (Dohrmann, '72; Sandler and

Tator, '76; Wagner et al., '78; Windle, '80; de la Torre,

'84; Beattie et al., '88; Fehlings and Tator, '88). The

progressive degeneration of spinal cord tissue is initiated

by a primary insult to neurons and vascular elements in the

region of the injury (Sandler and Tator, '76; Balentine and

Paris, '78a,b). The acute events then activate a cascade of

chemical reactions due to ischemia (Osterholm, '74), ionic

conductance changes (Young et al., '82; Stokes et al., '83),

and the accumulation of cytotoxic free oxygen radicals

(Demopoulos et al., '80) at the lesion site. Collectively,

these responses lead to the formation of a cavitated lesion

area within the spinal cord.

The chief functional consequences of this spinal lesion

are described in most basic neuroscience textbooks (e.g.,


Daube et al., '78). Briefly, the local degeneration of gray

matter leads to a loss of intrinsic and projection neurons at

the lesion site. This can result in the permanent

dysfunction of segmental actions, the extent of which is

dependent upon the size, type, and level of the resulting

lesion. The second outcome from the degeneration of spinal

cord tissue is the loss of functional continuity between

rostral and caudal levels of the spinal cord. Additional

compromise of autonomic and propriospinal systems underlie a

further level of complexity to the extent of functional

recovery (e.g. Cole, '88).

Recently, some emphasis has also been placed upon the

recognition of long-term functional effects at distant

regions of the central nervous system, as a result of

denervation and retrograde changes after axotomy (Beattie et

al., '88). Specifically, compensatory changes such as

collateral sprouting, receptor super-sensitivity, and

behavioral substitution, are likely to contribute to chronic

adaptive and maladaptive functions after spinal cord injury.

Treatments to Minimize Functional Loss

To date, most therapeutic approaches designed to spare

function or improve recovery after spinal cord injury have

been based upon the progressive pathology described above.

One such strategy includes pharmacological treatments aimed

at reducing or reversing the pathological process by

interfering with the cascade of secondary biochemical events

(reviewed in de la Torre, '80; Beattie et al., '88). For

example, corticosteroids have been used to stabilize membrane

changes and reduce edema after injury. Antioxidants have

been reported to promote functional recovery by reducing the

consequences of lipid peroxidation (Saunders et al., '87;

Anderson et al., '88). Other drugs have been employed to

improve functional characteristics attributed to axonal

transmission of long-tract fibers by altering ionic channel

conductances (Blight and Gruner, '87).

In addition to pharmacological approaches, surgical

procedures, including the removal of external compression

sources and stabilization of the vertebral segments

surrounding the injury site, are used to reduce the extent of

functional loss after injury (Ransohoff, '80).

Alternative efforts toward maximizing functional recovery

in later stages after injury have been directed at restoring

modulatory influences to spared regions of the spinal cord.

A number of animal experiments have suggested that the

circuitry below a lesion can be manipulated through the

application of pharmacological agents. Experiments using the

adrenergic agonist, clonidine, have demonstrated that

activation of catecholamine receptors can improve segmental

stepping behavior in cats in the first week after a spinal

injury (Forssberg and Grillner, '73). In addition,

application of bicuculline, a gamma-aminobutyric acid (GABA)

antagonist, has been associated with improvements in spinal


stepping after chronic spinal cord injuries. This latter

effect appears to be mediated by enhancing segmental

influences on the remaining circuits (Robinson and

Goldberger, '86). Conversely, the clinical use of the GABA

agonist, baclofen, has had beneficial effects in reducing

extreme spasticity in spinal cord injured patients by

enhancing the inhibitory influences upon segmental reflexes

(Bloch and Basbaum, '86). Thus, many proposed

pharmacological treatments for the chronic spinal injury

patient rely upon establishing a balance of receptor-mediated

neuronal activities at the segmental level.

Promotion of Axonal Regeneration or Sprouting

Over the past decade, advances in therapeutic strategies

have resulted in a progressive reduction of mortality with a

concurrent improvement in the quality of life following

spinal cord injury (e.g. Bloch and Basbaum, '86; Green and

Klose, '89). Nevertheless, a continued emphasis must be

placed upon research efforts to promote the repair of the

spinal cord and to restore functions normally mediated by

both segmental and long-tract systems.

Since neurogenesis is essentially lacking within the

adult mammalian central nervous system (CNS), there is no

mechanism for spontaneous replacement of neurons after

injury. Repair of the spinal cord is, therefore, dependent

upon the regeneration or sprouting of axons from existing

neurons. An emphasis in mammalian regeneration research has

been placed upon the evaluation of differences between the

peripheral nervous system (PNS), where injured axons

regenerate and are able to successfully reinnervate their

target tissues, and the CNS, where axons do not regenerate

such that they return to their original post-synaptic sites

(reviewed in Clemente, '64; Guth, '75; Bernstein et al., '78;

Kiernan, '79).

The aspects of these systems which have been contrasted

most often are the regenerative or sprouting capacity of the

axons and the permissive or inhibitory nature of their

environment. This interaction between axons and surrounding

cells received early attention by Ramon y Cajal ('28), who

observed small regenerative sprouts following experimental

injury to spinal cords of young kittens. The sprouts failed

to persist after two weeks, and no functional recovery was

observed. More recently, Aguayo and his colleagues confirmed

Ramon y Cajal's notion that the environment can profoundly

influence regeneration. By implanting pieces of peripheral

nerve into the brain and spinal cord, they demonstrated that

CNS neurons can extend and maintain long axonal sprouts

within the peripheral environment (David and Aguayo, '81;

Richardson et al., '82). Other researchers have provided

evidence of varying degrees of synaptic reorganization

following injury within the adult brain and spinal cord

(e.g., Cotman and Nieto-Sampedro, '85; Goldberger and Murray,

'88; Steward, '89b). Together, this work has inspired


renewed encouragement in the field of CNS and spinal cord


Strategies aimed at promoting the regeneration of spinal

cord axons include changes to the CNS microenvironment as

well as methods that might stimulate axonal sprouting and

elongation. Guth et al. ('85a) demonstrated that axons

within the spinal cord will not extend into a vacant lesion

site; instead, they must encounter a cellular terrain for

successful elongation. In addition, the establishment of a

dense meshwork of glial and connective tissue elements at the

lesion site may also present a problem for growing axons.

The concept of fibro-glial scarring as an impenetrable

barrier to elongating axons was championed by Ramon y Cajal

('28), and has been a topic of debate for the several decades

(reviewed in Reier et al., '83b; Reier and Houle, '88). With

this in mind, several approaches have been taken to prevent

or reduce the extent of glial/fibroblastic scar formation

after injury and thus promote axonal elongation. The

invasion of fibroblasts can be minimized by using closed

spinal cord injury models such as the weight-drop contusion

or clip compression approaches. In addition, pharmacological

agents have been applied to prevent the formation of scar

tissue at a lesion site (e.g., Windle et al., '52; Guth et

al., '85b). These results have suggested that axons may

extend a short distance into a spinal lesion where such a

scar is reduced.


Some recent investigations have been directed at

promoting the elongation of injured spinal cord axons using

different methods. One such approach involves the

implantation of cultured cells into a lesion (Siegal et al.,

'88; Wrathall et al., '89). These studies reflect a desire

to apply substances known to induce axonal elongation in

vitro to the injured spinal cord. The results suggest that

the effects may be more complex in vivo because of many

uncontrolled variables. Finally, the application of

electrical fields has also been investigated as a means to

increase the distance of axonal elongation (Borgens et al.,

'86, '87). These results, however, are inconclusive and

await further confirmation.

It is important to note that throughout the history of

spinal cord regeneration research, one recurrent difficulty

has been the unequivocal identification of regenerating axons

or collaterals into or across the site of a spinal cord

lesion. Using conventional histological procedures, such as

silver staining, the only way to verify that axons crossing

a lesion site represent true fiber growth has been to ensure

that the initial lesion constitutes a complete transaction.

This injury model, however, neither provides the most

conducive environment for regeneration nor represents a

clinical injury. Furthermore, even in the case of a complete

transaction, other important factors, such as the absolute

distance of fiber elongation, the origin and terminations of


axons, or patterns of reinnervation, cannot be verified using

these techniques.

Fetal Neural Transplants and Spinal Cord Repair

Reports of some degree of functional recovery following

transplantation of fetal CNS tissue into the brain (e.g.,

Gash et al., '80; Bjorklund and Stenevi, '84; Dunnett et al.,

'85; Buzsaki and Gage, '88) suggested that embryonic neural

tissue might promote neuronal repair following injury or

disease. These findings have led to the application of

similar transplantation strategies in the spinal cord. The

first intraspinal fetal grafting studies resulted in low

transplant survival rates as compared to similar experiments

in the brain (Nygren et al., '77; Patel and Bernstein, '83;

Das, '83; Commissiong, '84). Such difficulties served to

underscore the extreme pathological consequences of spinal

cord injury. It has been proposed that many of the grafts

failed to survive because they did not integrate with the

parenchyma of the injured spinal cord (Das, '83).

Following improvements in surgical procedures and careful

selection of donor tissue ages (Nornes et al., '83; Reier et

al., '83a; Reier, '85), more recent studies of intraspinal

transplantation have met with greater success. One approach

involves injecting suspensions of dissociated embryonic

brainstem cells caudal to the site of injury (Nygren et al.,

'77; Nornes et al., '83; Privat et al., '86). The focus of

this strategy is to restore supraspinal modulatory influences


to denervated regions below the level of a spinal lesion. An

emphasis has been placed upon the descending monoaminergic

systems which have been associated with the modulation of

segmental reflex and locomotor circuitries. These studies

have indicated that the injured spinal cord can be

reinnervated by grafted embryonic brainstem neurons.

Furthermore, such grafts can mediate some types of reflex

change after spinal injury or chemical denervation (Buchanan

and Nornes, '86; Moorman et al., '88; Privat et al., '86,


While transplants placed below a lesion site may

contribute to the replacement of modulatory influences,

recovery of sensation and voluntary motor capacities will

require applications that restore continuity at the lesion

site. Therefore, an alternative approach toward spinal cord

repair involves the transplantation of fetal tissue directly

into a lesion cavity (e.g. fetal spinal cord (FSC) grafts)

(Reier et al., '83a, '85,'86a; Houle and Reier, '88).

This approach differs from the transplantation of tissue

caudal to an injury, and it directly addresses three major

consequences of spinal cord injury. First, the presence of

embryonic tissue at the site of a lesion may provide trophic

influences to prevent degenerative changes after injury. For

example, fetal grafts placed into the injured spinal cord or

cortex of neonatal rats have been associated with a

significant reduction in the extent of cell death that is


characteristic of such lesions in the infant CNS (Bregman and

Reier, '86; Haun and Cunningham, '87). In addition, there

has been at least one suggestion that the presence of fetal

tissue in a spinal lesion cavity may prevent degeneration of

white matter fiber tracts in adult recipients (Das, '86).

Secondly, fetal tissue may serve to replace segmental

neurons at the level of the lesion. Interactions of

intrinsic and projection neurons may be important for the

repair of propriospinal influences after injury. This

approach for the replacement of damaged or diseased neurons

forms the basis for the transplantation of fetal neural

tissue into neurodegenerative and excitotoxin-induced lesions

in the brain.

The main objective of the intralesion grafting paradigm,

however, is to provide a neuronal framework that could

ultimately subserve functional integration of the rostral and

caudal spinal cord segments. In this context, the hypothesis

has been advanced that embryonic CNS tissue might be used to

promote spinal cord repair by providing a bridge for axons to

extend across the lesion (Nornes et al., '84; Reier, '85).

While results from recent studies suggest that descending

axons can extend across a FSC graft in newborn rats, there

is no evidence to date to indicate that injured CNS axons in

the adult will bridge a fetal graft to reinnervate their

original spinal cord target regions. However, an alternative

possibility is that transplants may establish a neuronal

relay pathway across a spinal cord lesion (Johnson and Bunge,

'83; Nornes et al., '84; Reier, '85, Reier et al.,'88;

Jakeman and Reier, '88).

Development of a Neural Relay Across a Spinal Injury Site

The concept of a neural relay has been discussed at its

most basic level by Shepard ('88). The three components of

the "synaptic triad" that form any neuronal circuit include

input neurons, intrinsic neurons, and projection neurons.

Complex variations of these components form the basis for

local circuits throughout the CNS (Rakic, '76). Several well

studied, integrative relays are found within the organization

of the dorsal and ventral horns of the normal spinal cord.

These circuits are responsible for the transmission and

integration of sensory and descending influences and

segmental reflex pathways.

The hypothesis of a relay with regard to FSC transplants

implies that the transmission of ascending and descending

information across a spinal cord lesion may be mediated

through interactions between host and graft tissues. These

interactions may take the form of afferent and efferent

projections between the surrounding host spinal gray matter

and fiber tracts and the intrinsic circuitry of the graft.

In the absence of axonal projections across the host-graft

interface, a relay circuit might be constructed by

interactions between axons which persist at the host-graft

interface and dendritic projections of host or grafted


neurons (Das, '83; Mahalik et al., '86; Clarke et al., '88b).

Alternatively, the relay may be more complex, as dictated by

differences in the relative growth and functions of different

host fibers. In the latter instance, the monoaminergic input

may serve to provide a modulatory influence upon the more

local host-graft interactions.

Experimental Goals

Despite efforts spanning over more than a decade of

research, the mechanisms that underlie various examples of

functional recovery following transplantation in the adult

brain are still unknown. Several recent review papers have

proposed a spectrum of possible mechanisms. In general, it

appears that functional changes following fetal neural

grafting may be obtained in a variety of ways in each model

system (Bjorklund et al., '87; Dunnett and Bjorklund, '87;

Gage and Buzsaki, '89). To better understand these models

and to test the capacity of the CNS for reorganization after

injury or disease, a recent emphasis has been placed upon

defining the anatomical correlates of host-graft

interactions. Through careful examination of the patterns of

axonal projections between transplants and the host CNS, the

strengths, potential mechanisms of behavioral improvement,

and the limitations of the grafting models can be assessed

more accurately.

Likewise, an important first step in determining a

potential functional role of FSC transplantation in the

spinal cord is to define the anatomical basis for integration

of host and graft tissues. Preliminary studies of

interactions between such transplants and the injured adult

rat spinal cord have indicated that some axonal projections

can form between the tissues (Reier et al., '85, '86a).

However, the purpose of these earlier studies was to identify

the general feasibility of transplantation and the

integration of such transplants into the adult spinal cord.

The objective of the following work is to identify, in

more detail, the anatomical basis for the role that FSC

transplants might play in the repair of the injured spinal

cord. The use of a variety of complementary neuroanatomical

methods will serve to define several aspects of host-graft

interactions, including the differentiation of regions in the

grafts and the development of axonal projections between

graft and host tissues. From these studies, information will

be obtained concerning the nature of neuronal relay

possibilities for transmission of information across the site

of a spinal cord lesion. In addition, the FSC transplant

model will be used to examine some of the biological issues

concerning axonal elongation within the adult spinal cord.



A number of spinal cord injury models have been used to

evaluate potential strategies for intervention and repair.

These include discrete lesions of specific fiber tracts,

chemical axotomy of fiber types, blunt contusion or

compression injuries, and complete or partial transaction

models (reviewed in de la Torre, '84; Beattie et al., '88;

Das, '89). The present investigations have employed a model

of transplantation into partial transaction cavities prepared

by aspiration immediately before grafting (acute lesions).

The transplantation methods used throughout these studies are

similar to those detailed in previous reports (Reier et

al.,'83a,'86a). Each experimental design has employed only

minor modifications of the procedures described below.

Experimental Animals

Adult, female, inbred Sprague-Dawley rats were used

throughout these studies. All of the rats were obtained from

Zivic-Miller Laboratories (Allison Park, PA) and weighed 200-

300 grams at the start of the experiments. The rats were

housed two per cage in the University of Florida animal

resources facility (accredited by the American Association of

Laboratory Animal Caretakers), according to the guidelines

established by the National Institutes of Health (Publication

number 85-23). They were examined daily by a veterinarian

and/or veterinary technician for general health conditions

and for any post-operative complications due to the spinal

cord injury. All surgical procedures were performed with

instrument preparation in anti-microbial benzalkonium

chloride solution (Zepharin HC1, Winthrop Breon Laboratories)

and 95% Ethanol.

Preparation of Lesion Cavities

The transplant recipients were anesthetized with

intramuscular injections of a mixture of ketamine (Ketaset,

60 80 mg/kg; Aveco Co.) and xylazine (Rompun, 10 mg/kg;

Mobay Corp.). The spinous processes were then exposed by

longitudinal incisions of the overlying musculature.

Bleeding from the superficial muscles was controlled with

epinephrine-impregnated cotton pellets (Gingipak, Belport

Co.,Inc.). A laminectomy was then performed at the

appropriate vertebral level using fine-tipped rongeurs, and

the host spinal cord was exposed by a longitudinal incision

of the dorsal meninges just lateral to midline. Using

iridectomy scissors and a glass micropipette with light

aspirative pressure, a cavity of 3 6 mm in length was

created in the parenchyma of the spinal cord, and bleeding

was controlled with small pellets of gelatin sponge (Gelfoam,

Upjohn Co.) soaked in a saline solution containing bovine

thrombin (Thrombostat, Parke Davis).


Preparation of Donor Tissue

For each transplantation session, a pregnant rat (E14,

embryonic day 14; E0 = day of insemination) was anesthetized

with 4.0% chloral hydrate (400 mg/kg, i.p.). A laparotomy

was performed and individual donor embryos (approximate

crown-rump length of 12 13 mm) were removed as needed and

placed into a standard tissue culture medium (Hank's Balanced

Salt Solution). The embryonic spinal cord tissue was then

dissected by removal of overlying skin layers and the

detachment of the spinal ganglia. The spinal cord was

stripped of embryonic meningeal layers and used for

transplantation within one hour of removal from the mother.

At the end of the transplantation session, the pregnant rat

was euthanized with an intracardiac injection of sodium

pentobarbital (Butler Co.).

Transplantation Procedure

Once hemostasis was achieved in the host, either one or

two pieces of donor spinal cord tissue were cut to

approximate the length of the prepared cavity. The donor

tissue and a small amount of tissue culture medium were

placed into the cavity with a flame-tipped micropipette.

After placement of the grafts, the excess tissue culture

medium was removed with light manual suction. A single

piece of hydrocephalus shunt film (Durafilm; Codman

Shurtleff, Inc.) was usually cut to a size just larger than

the lesion cavity and positioned directly above the graft.

The dural incision was closed with 10-0 interrupted sutures

and the spinal cord was then covered with a second piece of

synthetic dural covering. The overlying muscles were then

sutured in layers using 4-0 silk and the skin incision closed

with wound clips (Fisher).

Post-Operative Care

Following surgery, all rats received a subcutaneous

injection of long-acting penicillin (Dual-Pen; Tech America;

75,000 U.). They were carefully monitored and kept on a

heating pad or under a mild heat lamp until recovery from

anesthesia. Within 48 hours, the rats were returned to the

animal care facility where they were fed rat chow and water

ad libitum and maintained under a 12 hour light and 12 hour

dark schedule.

In the subsequent weeks after surgery, approximately 10%

of the animals initiated mild autotomy of the ipsilateral

forelimb or hindlimb (corresponding to lesion level) and were

treated with daily application of veterinary autophagic

repellent (Chewguard, Summit Hill Labs.). Those few rats

that failed to respond to such treatment within a few days

were sacrificed shortly thereafter and included in the data

analysis. This consideration contributed to the range of

post-graft survival times (post-graft intervals) referred to

throughout the work.




Intracerebral grafts of fetal CNS tissue have been shown

to compensate for a variety of functional deficits in

experimental animal models. This may occur through the

replacement of neuronal circuitries or neurotransmitters or

by the production of neuronotrophic substances within the

host brain (reviewed in Bjorklund and Stenevi, '84; Sladek

and Gash, '84; Bjorklund et al., '87; Dunnett and Bjorklund,

'87). The degree to which such grafts can subserve the

functions of the damaged brain region may depend upon the

ability of the grafted tissue to differentiate and integrate

with the host neural circuitry. Thus, several investigators

have examined the extent to which fetal neural transplants

will differentiate and exhibit organotypic characteristics

when placed into homotopic and heterotopic sites within the

CNS (e.g., Jaeger and Lund, '80; Alvarado-Mallart and Sotelo,

'82; Kromer et al., '83; Eriksdotter-Nilsson et al., '86;

Harvey et al., '88; Sorensen and Zimmer, '88a,b). From these

studies, it is clear that grafted regions of the CNS exhibit


different degrees of differentiation depending upon the age

of the embryonic donor tissue and the region from which it

is obtained.

In recent years, there has been some enthusiasm for the

application of fetal CNS tissue transplantation techniques to

the problem of spinal cord injury (e.g. Das, '83;

Commission, '84; Privat et al., '86; Reier et al.,'85, '86a;

Houle and Reier, '88). Together, these reports indicate that

intraspinal transplantation results in the survival and

integration of embryonic donor tissue in both neonatal and

adult recipients. One approach involves the introduction of

fetal neurons into the lesion site (Reier, '85; Reier et al.,

'85). In particular, homotopic grafts placed into the site

of a spinal lesion may serve as a source of specific

intraspinal neuronal populations with an inherent potential

for integrating with synaptic circuits above and below the

injury. Some degree of homotypic differentiation has been

indicated by studies in which fetal spinal cord transplants

were placed into intracerebral or intraspinal cavities (Reier

and Bregman, '83; Reier et al., '85, '86a,b). In these

initial investigations, distinct myelin-free regions were

observed in matured grafts, leading to the hypothesis that

these unmyelinated areas corresponded to the superficial

dorsal horn -- especially the substantial gelatinosa (SG) --

of the normal spinal cord. This region of the spinal cord is

easily identified in normal tissues based upon the paucity of

myelinated fibers, the small cells and compact nature of the

neuropil, and the density of peptide staining. Thus, the SG

provides a model system to determine whether grafts of FSC

tissue can differentiate and replace specific regions of the

normal spinal cord.

In the present study, the myelin-free regions of fetal

rat spinal cord grafts were examined in more depth to

determine the extent to which these areas develop cellular

and ultrastructural characteristics of the normal SG when

placed into lesion cavities in the adult rat spinal cord.

In addition, immunocytochemical methods were used to examine

the distribution of neurotensin-immunoreactive elements,

normally observed only within the superficial dorsal horn

(Portions of this study have been summarized in Reier et al,

'85; Jakeman et al., '89).

Materials and Methods

Animals and SurQical Procedures

Ten adult rats received intraspinal implants of E14 rat

spinal cord tissue. The surgical procedures were similar to

those described in previous reports (Reier et al., '83a,

'86a; Chapter 2, this volume). Each rat was anesthetized

with ketamine and xylazine and a laminectomy was performed at

the T13 vertebral (approximately L- I, spinal) level. An

aspirative lesion cavity, 3-4 mm in length, was created at

the exposed site. The lesion was then extended to include

either an extensive dorsal funiculotomy or lateral

hemisection. Whole segments of fetal spinal cord, 3-4 mm in

length, were dissected from donor embryos and introduced into

the lesion cavities (Reier et al., '83a).

Light and Electron Microscopy

At 1 to 2 months after transplantation, 5 recipients were

anesthetized with a lethal dose of sodium pentobarbital and

perfused through the heart with 0.9% NaCI followed by 5.0%

glutaraldehyde and 4.0% paraformaldehyde in 0.1M phosphate

buffer. Following the perfusion, segments of tissue

including the transplant and surrounding spinal cord were

then excised and divided into several transverse or

longitudinal slices. The specimens were subsequently

osmicated, dehydrated, and embedded in EM Bed 812 (Electron

Microscopy Sciences) for plastic thick sectioning. Regions

of the grafts classified as "SG-like" at the light

microscopic level were trimmed, and thin sections were

surveyed at the ultrastructural level using a Zeiss EM10C

electron microscope.


At similar intervals, the remaining intraspinal graft

recipients (n=5) were perfused with 4.0% paraformaldehyde in

0.1M Sorenson's phosphate buffer (pH 7.4). Tissue blocks

including the transplants were excised and prepared for

Vibratome (40 pm) sectioning. Adjacent 40 Am sections were

processed for the presence of myelin basic protein (MBP)- or

neurotensin (NT)- like immunoreactivity with the indirect


peroxidase anti-peroxidase (PAP) method (Sternberger, '76)

using primary antisera raised in rabbits. The antisera to

NT was obtained from Immunonuclear Corp. (Stillwater, MN.)

and the antiserum to MBP was provided by Dr. L. F. Eng (Palo

Alto, CA., VA Med Center). For PAP staining, tissue sections

were incubated overnight at room temperature in primary

antiserum diluted to 1:2000 with a solution of 0.3% Triton

X-100 in phosphate buffered saline (PBS). All incubations

were performed in the presence of 5% normal goat serum. The

sections were washed several times in PBS and then incubated

in goat anti-rabbit IgG (Cooper Biomedical or Sternberger-

Meyer) diluted 1:10 with the antiserum diluent for 45-60

minutes at room temperature. After a rinse with PBS, the

sections were incubated for 45 minutes in rabbit PAP (Cooper

Biomedical or Sternberger-Meyer) diluted 1:50. Following

several washes in PBS and 0.05M Tris buffer (pH 7.6), the

immunocytochemical reaction product was developed in a 0.05M

Tris buffer solution containing 0.05% 3,3'-diaminobenzidine

hydrochloride (DAB; Sigma) and 0.003% H202. Antibody

specificity was verified by the anatomical distribution of

immunoreactive elements in normal spinal cord tissue and by

the absence of immunoreactive elements when primary antibody

was replaced with normal serum alone.

Correspondence Between Plastic and Immunocytochemistry

In three specimens used for immunocytochemistry, 100 gm

vibratome sections were obtained to correspond to the

sections stained with antisera to MBP and NT. These sections

were osmicated and dehydrated as described above, and

embedded between vinyl slides with EM Bed 812. From these

sections, 2 Am sections were cut on an LKB ultramicrotome and

stained with 1.0 % Toluidine Blue. Similar 40 pm and 2 jm

sections were also obtained from the thoracic spinal cord of

2 normal rats to illustrate normal characteristics and

staining patterns within the substantial gelatinosa.

Light Microscopy and Cytological Characteristics

Sections of normal rat spinal cord, when stained with

antiserum directed against myelin basic protein (MBP),

exhibit a characteristic pattern of myelin distribution.

Specifically, the long fiber tracts appear densely stained,

while moderate staining reflects the presence of myelinated

fibers coursing throughout most of the central gray matter.

The most striking feature of this preparation, as also seen

with conventional myelin stains, is that region within the

dorsal horn which corresponds to the cytoarchitecturally

defined substantial gelatinosa (SG). This region stands out

against the background of intense myelin immunoreactivity due

to the paucity of myelinated axons (Fig. 3-1 a,b).

Fetal spinal cord transplants were stained with anti-MBP

to examine the differentiation of the grafts. All of the

transplants examined with this technique were heavily

1a) C C S?0 W0X
$14 4 k4J OC O0 4
0() 4 0 0 ) .
> I ...,. tP 0 0
-5 O ^ ,

4 J .,.

a) on 4 0 X,
V- > CO r-a (4
g & 0 r 0 41 k

40 4J

r.,t 0 41

.1 4 4- -g '. 4 J

a s44 H 0 t

p 1 H o

4 O.J V II

* .. O Ei'd

o 'C -
g o k g -*
0) 0 C '4 'a) 3 a)

0 C 0 4 44

I J 94 I

r4 C: 0 0 + 4
4 41 O-- --.C0
4- ) .,4 k rH
to V (0 401
. (iM H .,q 0




F .Al i,

.. ..I'

myelinated as indicated by areas of very dense staining. In

addition, large regions exhibiting moderate immuno-

reactivity were evident. Finally, the transplants usually

contained one or more areas which were conspicuous due to the

marked absence of MBP-like staining (Fig. 3-1 c,d). These

myelin-free areas typically assumed a convoluted configu-

ration within the grafts, and appeared as either single or

multiple patches or long strips of neuropil depending upon

the plane of section. In many cases, these regions were

located near the periphery of the grafts (Fig. 3-1 c);

however, some myelin-free areas were located more centrally.

Myelinated axons frequently curved along the surfaces of the

unstained regions, and often small bundles of anti-MBP

stained processes traversed the myelin-free zones in a radial

fashion, reminiscent of the pattern of myelinated primary

afferents projecting to deeper layers of the gray matter in

the normal spinal cord (Fig. 3-1 b,d).

With the perspective derived from MBP-stained grafts,

examination of toluidine blue-stained sections of FSC

transplants revealed areas that corresponded to patterns of

MBP- immunoreactivity (Fig. 3-2 a,c; Fig. 3-3 b). The 2 pm

sections contained regions of extensive myelination, as well

as numerous myelin-deficient areas within the graft tissue.

To determine whether the myelin-free areas identified

within the grafts by immunohistochemistry or toluidine blue

staining were indeed equivalent, adjacent sections were

Figure 3-2. Comparison of MBP and toluidine blue stained

a) Toluidine blue stained semi-thick section within an
intraspinal FSC transplant showing myelinated regions (m)
containing larger neurons, and unmyelinated patches (outlined
in arrowheads) with smaller neurons and processes.
b) MBP- stained section reveals a myelin-free region
(arrowheads) near the host-graft interface in another
recipient. Several regions of the graft lack myelin, yet this
region corresponds to an area containing small, tightly
packed neurons and processes in (c). Transplant (t) and host
gray matter (h16) are labeled.
c) Adjacent toluidine blue stained section from the same area
(within arrowheads) which is occupied by small cells and

Scale in a,c = 50 Am; b= 100 um.

llf. T ? f

L -

ait ,


- i

processed so that for each MBP-stained section there was a

corresponding 100 Am section embedded in plastic. In these

examples, zones that failed to show MBP immunoreactivity were

closely in register with homogenous unmyelinated areas in the

adjacent toluidine blue stained section (Fig. 3-2 b,c).

Further examination of these regions in plastic sections

revealed other similarities between the myelin-free areas of

FSC grafts and the normal SG laminae. As in the normal

substantial gelatinosa (Fig. 3-3 a,c), the unmyelinated zones

of matured transplants consisted of numerous small cells (7 -

15 Am) characterized by a thin rim of cytoplasm surrounding

a prominent nucleus. The nuclei of these cells, as of those

in the normal substantial gelatinosa, were round or oval and

often exhibited large clefts (cf., Fig. 3-3 c,d). These

cells were qualitatively distinct from the larger neurons

(14-50 Am diameter) that were found within the myelinated

regions of the transplants. In fact, the presence of larger

neurons within the unmyelinated areas was very rare. It was

interesting to note that some of the larger cells within the

myelinated regions were closely apposed to the border of the

unmyelinated zones. These general cellular relationships

were similar to the approximation of laminae III and IV

neurons with the normal SG.

In addition to features common to both the normal SG and

the myelin-free graft regions, there were some differences

r-H .0 ) CM H) r-
WH-4.JO4> rA0 r0
m rl 0C r-I 4J -r-I k
0 404o H 4.J 0
O .O> *eL 40
CE U 4-4 3
4J 0 tO 4J*-H Wo OH
go 0 0 to r .,. to
M r-44 X0 O 0 'd 4J
q ($4 g 41
O1 0
4 .0 toSo54 g
t H E o E 0 0 4. )

4) 4) 144 *0
4- f *44J0 44

Ch 4.H0. 40 0
-r. 4EU t CH c

5~ &> *M .^
0 O 4)Q4) -4 Fi

M r- 4. 4c 0 0

(1 14~4 `4 44 EUZU
E- U*4 1 0 g
4Ji r- 4) 0a w O., ^ii4

4UU)0 4J H

9 :|i l ^;O

t r E 4- .0EU 0.4

O 0 :0 0o 0 rE oIV
to to 04to 41 Ul 4'- )n 4

S ) C O 4 C to m
> ~ ~ 0 to ra IHIr


* 1. M .,CHO* 4 )o 0 O
I eag ,g ji 4 *H c 4 r 4J
0 -(1)EU 0 r.

.9-4 j4 0

90 N 0A (j 4 9

4> c004000U)9 C- )
a (dC 4J^3M +~ rU 0) r. 4
0n ( +j Mj to --4 a) V ^ (

0 0 ) 0r-to r-U 0
.0 (4J e M OlB '0 0 4 0
W J 4j4) J4~4i ) 10HU
04 4J4EU4 Q). 0 E
44 V94 0u OC

04 -0--UH4 0-a)0 to 0
>4 4 to to :1 W 4.jQ 4J-'



between the two regions. This was particularly evident with

regard to the distribution of cells and neuritic processes.

In the intact spinal cord, SG neurons can be differentiated

into two layers (i.e., IIi and IIo; Ralston, '79;

Szentagothai, '64a). However, no obvious cytoarchitectural

lamination was seen in these regions of the transplants. In

addition, many small, circular neuritic profiles were seen in

transverse sections of the normal SG, thus reflecting their

orientation parallel to the longitudinal axis of the spinal

cord (Fig. 3-3 c). In contrast, the neuritic processes in

the myelin-free areas of the grafts seemed more randomly

organized, and sectioned profiles assumed many orientations.

Electron Microscopy

Neurons within the normal substantial gelatinosa were

generally spheroid (Fig. 3-4 a) or fusiform (Fig. 3-4 b) in

shape. The perikarya ranged from 8 20 pm in diameter, and,

as seen with the light microscope, they usually contained a

large nucleus within a narrow rim of cytoplasm. These cells

were embedded in a neuropil that primarily consisted of

tightly packed, small unmyelinated axons and small to

intermediate-sized dendrites. The axons were often organized

into bundles or fascicles which were more evident when the

tissue sections were cut in the transverse plane. Many of

the synapses identified within the SG were axo-dendritic

in nature, although other synaptic types were found. In

Figure 3-4. Ultrastructure of the normal substantial

a) Transverse section contains a neuronal cell body (n) and
the compact neuropil containing abundant axo-dendritic
synapses (ad) interspersed with longitudinal bundles of
unmyelinated processes (arrows). Large glomerular axonal
processes were often observed (star).
b) Oblique section through the SG region of another normal
specimen. The fascicles are more difficult to discern than in
(a) .

Scale in a,b=2.5 Am.



A *'



'. VI l

S 1F .I


addition, large glomerular complexes were evident in the

substantial gelatinosa of these normal sections as well.

A survey of the unmyelinated regions within FSC

transplants in low power electron micrographs revealed many

characteristics similar to the normal SG (Fig. 3-5). The

neurons in these areas were also small and contained large

nuclei with prominent indentations. The cells were closely

spaced, and were surrounded by compact neuropil consisting of

small axons (0.1 0.3 gm) and intermediate-sized dendrites

(0.4 1.6 gm). Occasionally, small bundles of unmyelinated

processes were seen which resembled the fascicles in the

normal SG. However, many of the axons and dendrites were

more randomly oriented (Fig. 3-5, 3-6). Except for an

occasional swollen neuritic profile containing lysosomes and

degenerating mitochondria, axonal and dendritic processes

did not display irregular cytological characteristics. In

some transplants, hypertrophic astrocytic processes were

observed particularly near the periphery of the grafts (Fig.

3-5, 3-6).

Many of synaptic contacts observed within the

unmyelinated graft regions were axo-dendritic, with axo-

axonic synapses occasionally being present as well. In

addition, a rare axo-somatic synapse could be found in the

normal and graft myelin-free areas. The boutons within the

myelin-free regions usually contained aggregates of small,

agranular vesicles (Fig. 3-6). Usually, the vesicles were

Figure 3-5. Low power electron micrograph of an SG-like
region in an E14 intracerebral transplant.

The small neurons in these regions were similar to the normal
SG regions in size, shape, and the presence of nuclear
clefts. An axo-dendritic (ad) and axo-somatic (as) synapse
are shown. While some unmyelinated axons traveled in
fascicles (arrows), most axonal and dendritic processes
assumed a variety of orientations, and appeared to lack some
of the organization of the normal spinal cord. Large
filamentous glial processes (*) were sometimes seen in these

Scale = 2.0 Am.

Ar.. ** <


Figure 3-6. Higher magnification of SG-like regions in an
intraspinal transplant.

The vast majority of synapses were axo-dendritic (ad), and
vesicles were usually clear (agranular) and round (arrows),
although flattened vesicles and dense-cored vesicles
(arrowheads) were also observed. Astrocytic processes were
sometimes present (*).

Scale = 0.5 gm.

* ..

1o 6Ulla

..a ~. .1

0.* ;.
, -

74 .

if~ -
4' s
-s *

- S


~ ~

* '

* .w
mqF.. ~
,q.j~jy -
S.. -a--



r :",-I

round, rather than flattened, and some terminals contained

small dense-cored vesicles. Similar presynaptic structures

were also seen in the normal SG. However, the scalloped

terminals and related glomeruli characteristic of normal

primary afferent innervation of the SG were not found within

the grafts examined in this study.

Neurotensin-like Immunoreactivity

When sections of normal spinal cord were reacted with

antisera to neurotensin (NT), staining was restricted to the

lamina II region of the superficial dorsal horn, where it was

seen in two distinct layers of very fine processes (Fig. 3-

7 a; Seybold and Elde, '82). Labeled axonal profiles were

not found in any other region of the normal spinal cord, and

no immunoreactive cell bodies were found in these sections.

Staining of FSC transplants with antisera raised against

NT revealed regions throughout the grafts which contained

small immunoreactive processes. Many of these patches

corresponded with myelin-free regions in neighboring sections

stained with anti-MBP (Fig. 3-7 b,d). However, in contrast

to the normal spinal cord, NT staining within the transplants

did not form two distinct bands. Other differences were also

observed between the patterns of NT staining and the normal

SG. In addition to the patches of fibers that corresponded

to MBP-free regions of the grafts, some NT fibers were

distributed throughout the myelinated areas of the

transplants. Furthermore, NT-like cells were also found

H Cr 4.1 a) >i) M EU
0 E-4 V 40) : X V X *-- .C
. Z 0 ) A. E-4 -4. 41 X
-4 ,1 *0C 9
a 4) 4 .4

O g C g00 g

3r 0 to ,i ..... !
9.4) U) = 4) 0 0

t0 p24 0I W 4 -
0 1 ) .:"

to- ^ 14 3,.O

H o o
) H r 0

H L 4 c. *q c

c 4'"- OC C O-
4.) 0*v *3 0 9 0

-v. 0. CO 4 U)
to P|||41 4J

0 r. a|co$4 0 $4 o
r 4 C 0 *n 0
C 0

p -.'- 0)O
* 1 0 04-4 x H

44 0V 400"- f

40 040 -0 0 >MVO((% .4

0-10.v.4,QO H Ic
)) t o 4 g -a ) (, u ) v a) -.4 o

C 0 1 C0 CUO40 (0 4-.) 4C 0
r4 5 400 )00 0)
E'0, 4 rCO Z4- 0
& 0 W 0Q0) 0 t


..~' .-.
"~~ ~ W.,. .; d

within the grafts (Fig. 3-7 b). These cells were small and

multipolar in shape, and were often found in groups of two or


Neurotensin immunoreactive fibers were often observed at

the periphery of the transplants. Interestingly, in some

examples, these labeled fibers could be followed across the

host-graft interface, where they appeared to innervate

ventral regions of the host gray matter. This region does

not normally contain neurotensin fiber ingrowth.

In previous descriptions of matured intracerebral and

intraspinal transplants of FSC tissue, regions of graft

neuropil were identified based upon the absence of myelinated

fibers (Reier et al., '85, '86b). The indication that these

regions might reflect some organotypic differentiation was

examined further in an experiment in which pregnant rats were

injected with tritiated thymidine on either day E12 or E14.

Labeled donor tissue was removed from fetuses in one uterine

horn and transplanted; fetuses in the contralateral horn were

left to complete gestation (Reier et al., '83a). At one

month post-transplantation, autoradiography indicated that

the nuclei of neurons in the myelin-free regions of these

transplants exhibited the same relative degree of labeling as

did the nuclei of those cells present in the superficial

laminae of the intact spinal cords of the littermates of the

donor fetuses (Reier et al., '83a,'86b).

It thus appeared that the myelin-deficient regions in the

fetal spinal cord grafts were the counterparts of the normal

substantial gelatinosa. These general criteria, though

intriguing, did not provide sufficient proof of the exact

nature of the myelin-free areas in the grafts. In this

context, it is well recognized that other characteristics

make this region distinct from the rest of the gray matter in

the intact spinal cord. In particular, the abundance of

small cells led Rexed ('52) to make the distinction of lamina

II of the cat spinal cord, and similar studies have shown

this region within the rat spinal cord as well (Molander et

al., '89). Examination of the ultrastructure of this region

has also served to define the types of processes and synaptic

profiles that distinguish the substantial gelatinosa (lamina

II) from the surrounding marginal layer (lamina I) and lamina

III (Ralston, '79; Szentagothai, '64a). Additional

identifying features have been noted through the use of

immunocytochemical techniques, which have demonstrated a

variety of peptide-containing cells and fibers (reviewed in

Seybold and Elde, '80; Gibson et al., '81; Hunt, '83;

LaMotte, '86). With these characteristic cytological and

immunocytochemical features as a basis for comparison, the

present study has provided additional evidence in support of

the organotypic differentiation of substantial gelatinosa-like

regions in transplants of fetal spinal cord tissue.

Myelin-free Areas and Peptidergic Elements

Previous studies have demonstrated regions of dense

immunoreactivity obtained with antibodies to several peptides

which are normally associated with the substantial gelatinosa

(Reier and Bregman, '83; Reier et al., '85). Patches of

tissue within the transplants stained heavily with antibodies

to met- and leu-enkephalin, somatostatin, and substance P.

These patches have been identified within FSC transplants

placed into the adult brain or neonatal and adult spinal

cord. In addition, similar findings have been reported

regarding the differentiation of regions of dense peptide

staining in FSC transplants that develop in oculo (Henschen

et al., '88).

In this study, a comparison has been made between the

myelin-free regions of FSC transplants and the SG of the

normal spinal cord. Standard light and electron microscopic

observations revealed many similarities between these areas,

especially with regard to cell and process sizes and the

compact nature of the neuropil. With regard to peptide

staining patterns, an emphasis was placed upon the

distribution of NT- containing fibers within the transplants.

As described for other peptides, regions within the FSC

transplants contained distinct patches of NT-like fibers that

corresponded to unmyelinated regions of the graft. In the

same spinal cord tissue rostral and caudal to the grafts,

however, NT-containing axons were found only within the

substantial gelatinosa region of the spinal cord.

Developmental Implications

The observation that fetal spinal cord tissue can exhibit

some degree of organotypic development is consistent with

reports that transplants of tissue from various embryonic

brain regions can achieve cytoarchitectural and

ultrastructural characteristics corresponding to those of the

homologous areas in the intact CNS (e.g., Kromer et al., '79,

'83; Lund and Harvey, '81). Fetal spinal cord tissue has

also been shown to exhibit some cytoarchitectural or

immunocytochemical characteristics resembling the normal

dorsal horn when grown in tissue culture (Naftchi et al.,

'81; Sobkowicz et al., '68) or in oculo (Henschen et al.,

'85). The present findings show that the differentiation of

the myelin-free areas also occurs when FSC tissue grafts

develop within the adult spinal cord.

In related studies currently in progress, similar

unmyelinated regions have been observed in cell suspension

grafts of rat FSC within the contused rat spinal cord

(Winialski et al., '89) and in solid piece grafts of cat

fetal spinal cord tissue in the adult cat spinal cord

(Anderson et al., '89; Reier et al., in preparation). These

regions have thus far been identified on the basis of 2 gm

thick plastic sections and anti-MBP stained sections of the

grafts. Additional observations suggest that the myelin-free

areas may also be analogous to small fiber-free areas

revealed with immunocytochemical staining using polyclonal

antibodies raised against the phosphorylated form of the

heavy neurofilament protein (see silver staining patterns in

Lund and Harvey, '81). However, one recent report has

indicated that SG-like regions are not seen following

injections of cell suspensions from E12 E13 fetal rat spinal

cord into the ibotenic acid lesions of the lumbar spinal cord

(Nothias et al., '89). This difference may reflect the

different donor ages used (Kromer et al., '83), however, SG-

like areas have been identified in grafts of E12 rat spinal

cord tissue placed into the rat brain (Reier et al., '83a).

Thus, while the procedures used in this study were not

applied by Nothias et al., it may be possible to observe SG-

like regions within their grafts placed into ibotenic acid

lesions by utilizing specific markers such as the antisera

raised against dorsal horn peptides or MBP.

The presence of a dorsal horn component in FSC grafts is

likely to be related to the developmental timing of this

region of the spinal cord. Evaluations of spinal cord

histogenesis in the rat (Alvardo-Mallart and Sotelo, '82;

Nornes et al., '74) have indicated that there is a peak at

approximately E15-E16 in the generation of neurons which

ultimately comprise the dorsal horn. These cells then

migrate with the majority of neuroblasts reaching the

presumptive dorsal horn region two days later. The

maturation of the SG-like areas in these fetal spinal cord

grafts must therefore occur after transplantation since donor

tissue in these experiments was obtained at E14-E15. These

considerations pertaining to cell birthdates and onset of

migration also suggest that the clustering of small neurons

into the SG-like regions may be due to the persistence of

intrinsic recognition cues which influence the aggregation of

these cells during normal development (see also Kromer et

al., '83). Related studies have also shown that these

intrinsic cues are also retained if the graft is placed into

heterotopic sites, or when it is entirely isolated from host

afferent inputs by transplanting the fetal spinal cord with

the surrounding meninges attached (Jakeman et al., '89).

Anomalous Features of the Substantia Gelatinosa-Like Regions

While many features of the myelin-free regions of FSC

grafts reflect a homology with the normal SG, it is clear

that the correspondence was not perfect. Several aspects of

these areas represent a departure from the normal

organization of the mature superficial dorsal horn. For

example, the myelin-free graft regions lack the precise

orientation and formation of a dorsolateral cap shape with

cells organized in discreet layers. In addition, the

definition between outer and inner layers of lamina II

observed in the normal SG was not observed in either 2 pm

plastic sections or with by immunocytochemical staining with

antiserum to NT. Finally, while these graft regions

contained some bundles of unmyelinated axons, the parallel

longitudinal arrangement of neuronal processes characteristic

of the normal dorsal horn was absent from the myelin-free

regions of the grafts.

It is likely that some of these differences are related

to specific aspects of the grafting procedure, such as the

initial orientation of the graft tissue, donor age, and

changes in the precise timing of developmental cues (Kromer

et al., '83; Stenevi et al., '76). In addition, the

topography of some of these transplants may be distorted by

spatial restraints, which can contribute to the

organizational differences observed. The differentiation of

organotypic regions within an abnormal cytoarchitectural

framework has also been observed in other types of fetal CNS

transplants (e.g. Kromer et al., '83; Mufson et al., '87;

Sorensen and Zimmer, '88b).

Another factor that may contribute to the atypical

features observed here is the relatively deafferented state

of the transplant. As noted in our electron microscopic

results, the SG-like regions in these grafts lacked the

synaptic terminals characteristic of primary afferent

innervation. A qualitative analysis of the synaptic

composition of these grafts revealed similarities in

ultrastructure to the deafferented dorsal horn, specifically

in the absence of organized glomerular complexes (Rethelyi

and Szentagothai, '69; Coimbra et al., '74). While such

complexes have been found in dorsal regions of FSC grafts

intentionally innervated by primary afferent fibers (Itoh

and Tessler, '88), SG-like areas deeper within the

transplants receive few such afferents.

Innervation of FSC transplants from adult host fibers is

largely restricted to the periphery of the grafts (Chapter

4). Thus, the developing grafts may also be lacking much of

the descending modulation present in the normal spinal cord.

Therefore, rather than being indicative of aberrant

development, the atypical features recorded may instead

reflect the normal development of SG-regions in circumstances

of decreased afferent input. Such a situation would not only

alter the patterns of migration and lead to cytoarchitectural

differences, but might also prevent the normal expression of

neuropeptides and neurotransmitters.

In the present study, differences were observed between

the pattern of neurotensin-like immunoreactivity in the FSC

grafts and in the normal SG. Specifically, NT-positive cells

and fibers were observed throughout the grafts, while only

fibers were found in the normal spinal cord, and those were

restricted to the SG region. However, application of

intraventricular colchicine treatment and better fixation

methods have been used to identify NT-containing cells and

fibers in the normal spinal cord. These studies have

indicated that NT-like cells can be found in lamina I and V-

VII in addition to the SG in normal rats (Gibson et al., '81;

Seybold and Elde, '82; Miller and Seybold, '87). Thus, the

presence of these cells throughout FSC transplants in the

absence of such treatments may reflect alterations in peptide

expression or axonal transport mechanisms.

Similar discrepancies with regard to peptide expression

have been found recently in cortical and spinal grafts that

developed in oculo (Eriksdotter-Nilsson et al., '87; Henschen

et al., '88). Taken at face value, the results seen in oculo

suggested that the disturbed patterns of peptide staining

might be attributed to the isolation of these transplants

from the environment of the CNS. However, our present

findings indicate that some alterations in peptide expression

exist even in well integrated grafts that develop within

homotopic locations in the CNS. While these grafts contain

some afferent ingrowth from descending fibers, such input is

limited, and may not be sufficient to induce the normal

expression of such peptides. Further support for this

hypothesis comes from recent studies which have shown

differences in calcitonin gene-related peptide (CGRP)

immunoreactivity of hindlimb motoneurons after chronic spinal

cord transaction (Arvidsson et al., '89).

Implications for Repair

The functional role of cells in the mature SG of the

spinal cord is a topic still under intense investigation

(reviewed in Willis and Coggeshall, '78; Cervero and Iggo,

'80). The termination of unmyelinated primary afferent

fibers in this region provides the basis for theories

concerning its role in the modulation and gating of pain and

reflex functions (Melzack and Wall, '65; Willis and

Coggeshall, '78). Szentagothai ('64a) used Golgi stains and

degeneration techniques to reveal the morphology and

projection patterns of SG neurons and proposed that the SG is

primarily a closed system, dominated by local projection

neurons that extend no more than 2-3 segments. More recent

evidence obtained with axonal tracing techniques has shown

that at least some of these SG neurons can project as far as

the medulla (Giesler et al., '78) and thalamus (Willis et

al., '78). It is now known that axonal projections from the

SG also extend into deeper laminae of the spinal cord as well

(Light and Kavookjian, '88). Therefore, the differentiation

of SG areas within FSC transplants suggests a source of

intrinsic modulatory cells as well as some projection neurons

that may play a role in the formation of a neural relay for

somatosensory information.

The availability of specific characteristics and peptide

markers to identify the substantial gelatinosa of the normal

spinal cord has allowed the identification of patches of SG-

like regions within FSC transplants. While these areas may

represent only a small portion of the total circuitry of the

spinal cord, it is likely that other regions of the embryonic

spinal tissue also differentiate and exhibit characteristics

of homotopic areas. Further anatomical markers for the

intermediate and ventral regions of the spinal cord may be

used to address this issue. However, one feature of the

normal spinal gray matter that is rarely seen in these

transplants obtained from E14 rat embryos is the development

of groups of large motoneurons within the grafts. Greater

numbers of motoneurons have been identified in grafts derived

from younger (E12) embryos (Reier et al., '83a) Thus,

selection of different aged donor tissue may be useful for

enriching transplants in either ventral or dorsal neuropil

(e.g. Nothias et al., '89).

It is not known which region of the embryonic neuraxis

is best suited to restore function in the injured spinal

cord. However, successful repair of damaged neural networks

may require the reconstruction of certain suprasegmental and

intraspinal circuits. Recent studies have shown that in some

instances, homotopic grafts are innervated in varying degrees

by host serotonergic and primary afferent fibers (Bregman

'87; Reier et al., '85, '86a; Tessler et al., '88). Both of

these axonal systems, as well as many other identifiable

fiber populations, normally project to the SG. Because these

SG-like areas are easily identified within FSC transplants,

and because the afferent innervation of the normal SG is well

characterized, this transplantation model should provide a

valuable opportunity for testing the ability of host axons to

recognize regions of these grafts with similar cytological

and peptidergic characteristics to the SG of the normal

spinal cord. Such information can be useful in further

understanding the potential of the grafts to reconstruct

specific circuitries in the injured spinal cord.

The differentiation of at least one region of the normal

spinal cord within FSC grafts suggests that these grafts may

replace populations of intrinsic spinal cord neurons. The

following studies are designed to determine whether these

neurons form projections both within the grafts and between

the transplant and host spinal cord.


In neonatal rats, transplants of fetal spinal cord (FSC)

tissue have been shown to provide an environment conducive to

the elongation of some descending axons through a spinal

injury site (Bregman, '87). In addition, when placed into

hemisection lesions in these newborn rats, FSC transplants

have been shown to improve the development of specific

aspects of hindlimb function as compared with rats with

hemisections only (Kunkel-Bagden and Bregman, '89).

In adult rats, however, there is no evidence to support

the concept of axonal regeneration of descending axons across

fetal spinal cord transplants. Nevertheless, the propagation

of some aspects of ascending and descending information might

be achieved by fetal grafts through the establishment of a

neuronal relay between the rostral and caudal regions of the

recipient spinal cord (Reier et al., '85, '88).

To test this hypothesis, the present study was designed

to identify and characterize patterns of axonal interaction

established between FSC grafts and adjacent regions of the

host spinal cord. Therefore, the purpose of the first

experiment was to extend preliminary WGA-HRP tracing studies

which had suggested some axonal integration between host and

graft (Reier et al., '86a). A fluorescent retrograde tracer

(Fluoro-Gold) was then used to determine the distribution of

cells contributing to axonal interactions. To complete the

axonal tracing studies, the anterograde transport of the

plant lectin Phaseolus vulgaris leucoagglutinin (PHA-L) was

included to reveal the patterns of the axonal projections

from local host and graft neurons and their relationship to

the host-graft interface. The combination of these three

contemporary axonal tracing techniques offers a unique

approach toward evaluating the local interactions between FSC

transplants and the adjacent regions of the spinal cord.

Complementary information about the ingrowth of specific

populations of host afferents into the transplants was then

obtained in a second group of animals by immunocytochemical

staining of transplant sections with antisera raised against

serotonin (5-HT), oxytocin (Ox), tyrosine hydroxylase (TH)

and calcitonin gene-related peptide (CGRP). Finally,

additional sections from both axonal tracing and

immunocytochemical specimens were stained with antiserum

against glial fibrillary acidic protein (GFAP) to examine the

relationship between host-graft projections across the

interface and the patterns of glial reactivity. Portions of

this study have been summarized previously (Jakeman and

Reier, '88).

Materials and Methods

Animals and Transplantation Surgery

A total of 99 female, adult rats received transplants of

FSC tissue according to a modification of previously

described methods (Reier et al., '83a, '86a; Chapter 2).

Each transplant recipient was anesthetized with ketamine and

xylazine, a laminectomy was performed at the T13 vertebral

level, and a cavity of 3 6 mm in length was created in the

left half of the spinal cord. For this study, the lesion was

routinely extended to a full hemisection by removal of both

the lateral and ventral columns. The overlying dorsal roots

were reflected laterally during preparation of the cavity,

and replaced after grafting. Although no effort was made to

intentionally sever or remove the rootlets (Tessler et al.,

'88; Houle and Reier, '89), they were sometimes injured

during the surgical procedure. Once hemostasis was achieved

in the host, the donor tissue was placed into the cavity and

the dura and superficial tissues were closed in layers.

Axonal Tracer Application and Tissue Processing

At post-graft intervals ranging from 6 weeks to 14

months, transplant recipients were re-anesthetized with

ketamine and xylazine and prepared for tracer application.

The region containing the graft and surrounding host spinal

cord was exposed by removing new bone growth and extending

the original laminectomy rostrocaudally. After the tracer

was injected, the surface of the cord was washed with

physiological saline and a drop of mineral oil was placed

over the spinal cord to minimize diffusion of the tracer into

the surrounding tissues. The spinal cord was then covered

with a piece of Durafilm and the wound was closed as

described above.

Horseradish Peroxidase (HRP) and Wheat Germ Acqlutinin HRP
conjugate (WGA-HRP)

Anterograde and retrograde labeling. A combination of

HRP (Type VI) and WGA-HRP (Sigma Chemical) were used for both

retrograde labeling of cells and anterograde filling of axons

(Mesulum, '82), as described in previous studies. In the

first part of the experiment, mixtures of the two tracers

were applied to transplants (n=16) using a variety of

techniques: (a) Seven rats received pressure injections of a

solution of 20% HRP and 1.0% WGA-HRP using a 1.0 &l Hamilton

syringe or a nitrogen burst picospritzer (Reier et al.,

'86a); (b) Two rats received iontophoretic injections of 2%

WGA-HRP; (c) one rat received a pledget of Gelfoam soaked in

20% HRP and 2.0% WGA-HRP; and (d) HRP and WGA-HRP were

applied to the remaining six rats using crystals dissolved

onto the end of a tungsten wire (Houle and Reier, '88).

For the reciprocal study, HRP and WGA-HRP were applied

to the host spinal cord with a tungsten wire ((d) above,

n=ll). In order to examine the HRP transport characteristics

using this method, a similar tungsten wire was placed into

one normal rat at the T13 vertebral level.

HRP histochemistry. After allowing 48 72 hours for

transport of the tracer, the recipients containing HRP and/or

WGA-HRP injections were deeply anesthetized with sodium

pentobarbital and perfused transcardially with 150 ml

heparinized 0.9% NaCIl followed by 250 ml fixative (1.0%

paraformaldehyde + 2.5% glutaraldehyde in 0.1 M Sorenson's

phosphate buffer). Tissue blocks, including the transplant

and 5.0 10.0 mm of host spinal cord rostral and caudal to

the graft, were removed. Vibratome sections (50 Mm) were cut

in the sagittal or horizontal plane. The sections were

reacted within 2-4 hours according to the tetramethyl-

benzidine (TMB) protocol of de Olmos et al. ('78). Sections

were then mounted onto gelatin-coated slides and selected

slides were counterstained with 1.0% Neutral Red to reveal

the cytoarchitecture of the host and graft tissues.

Fluoro-Gold (FG)

Retrograde labeling. For the second set of experiments,

a 2.0% solution of FG (Fluorochrome, Inc.; Englewood, CO) was

made in 0.9% NaCl and the solution was then drawn into glass

pipettes (40 50 Am tip diameter). After a dural incision

was made with the beveled end of a 25 gauge needle, the FG

solution was injected into the transplants (n=12) or the host

spinal cord (n=19) using a rapid nitrogen burst (Pico-

spritzer) applied to the end of the glass micropipette. The

approximate injected volume of the FG solution was estimated

by measuring the diameter of the "hemisphere" ejected onto a

parafilm sheet and using the approximation (v= (2r(d/2)3)

/3). Volumes of 0.1-1.0 Al were injected into the

transplants and 0.5-1.5 Al into the host tissue. One normal

rat also received an injection of 0.5 pl of FG solution at

the T13 vertebral level for comparison.

Tissue processing. The FG- containing tissue was

processed as described by Schmued and Fallon ('86). At 4

days after the injection, the rats were perfused with saline

followed by fixative containing 4.0% paraformaldehyde and

0.25% glutaraldehyde in 0.1 M phosphate buffer. Spinal cord

blocks containing the transplant and 4.0 mm of the

surrounding rostral and caudal spinal cord were removed and

postfixed in the same fixative for 2 hrs to overnight at 40C.

Vibratome sections of 40 Am were cut in the sagittal plane.

In addition, every sixth section was saved in 0.1 M PBS for

subsequent immunocytochemical detection of GFAP (see below).

In 4 of the rats that received larger injections of FG

into the transplants, six additional tissue blocks were also

sectioned. These included cross sections of the host spinal

cord 4 6 mm rostral and caudal to the transplant,

horizontal sections from cervical and thoracic spinal cord,

and sections of host brainstem and brain. The dorsal root

ganglia from these recipients were embedded in paraffin and

sectioned at 15 Mm. The Vibratome and paraffin sections were

mounted directly onto gelatin-coated slides. Slides from

paraffin blocks were heated to 370C for 12 hours, and

deparaffinized. All FG slides were cleared in xylene,

coverslipped with Fluoromount (Gurr Bio/medical Specialties;

Santa Monica, CA), and viewed on a Zeiss Axiophot microscope

with fluorescent UV illumination.

Phaseolus vulgaris leucoaqqlutinin (PHA-L)

Anterograde labeling and tissue sectioning. To examine

the patterns of axonal elongation into graft and host

tissues, anterogradely-filled axons were identified by

immunocytochemical detection of PHA-L (Vector Laboratories

Inc.; Burlingham, CA). The tracer was applied by a

modification of the methods of Gerfen and Sawchenko ('84).

The PHA-L was dissolved to 2.5% in 10 mM phosphate buffer (pH

8.0). Glass micropipettes were cleaned with acetone and 100%

ethanol and broken to a tip diameter of 10 15gm. After

exposing the transplant (n= 13) or host spinal cord (n=8),

the tracer was applied to the appropriate site by

iontophoresis for 20 minutes using a 5 MA interrupted

positive current (7 sec on, 7 sec off).

After allowing 7 17 days for transport of the PHA-L,

the recipients were perfused as above with fixative

containing 4.0% paraformaldehyde and 0.25% glutaraldehyde.

The cord blocks including the transplant and 4 7 mm of the

surrounding rostral and caudal spinal cord were removed and

postfixed overnight at 4C. Sagittal sections of these

blocks were cut at 40 Am on a Vibratome and stored in 0.02 M

potassium phosphate buffered saline (KPBS). Every sixth

section was saved for immunocytochemical staining with

antibodies to GFAP (see below).

Immunocytochemical detection of PHA-L. The remaining

free-floating Vibratome sections were processed for the

identification of cells and processes containing PHA-L. The

sections were first washed in 0.02 M KPBS and incubated for

2 4 hrs in a preblocking bath containing 2.0% normal rabbit

serum and 0.3% Triton X-100. All the sections were then

incubated in goat anti-PHA-L (Vector) diluted 1:5000 in KPBS

for 36 hours at 4C and 2 additional hours at room temper-

ature. The sections were rewashed and then processed with

biotinylated rabbit anti-goat IgG (1:225) and Vector Avidin-

Biotin-peroxidase Complex (ABC) as per the supplier's

instructions. The final peroxidase conjugate was reacted

with H202 in the presence of 0.005% DAB. The DAB reaction

was done either with the addition of 0.125% nickel ammonium

sulfate (black reaction product) or in the absence of nickel

(brown reaction product). The nickel-enhanced sections were

counterstained with 0.1% Cresyl Violet or 1.0% Neutral Red

prior to coverslipping.

Histological analysis of anatomical tracers

Mounted serial sections containing the tracer injection

sites were examined to determine the location of the

injection site and the extent of tracer diffusion relative to

the host-graft interface. Each specimen was then accepted or

rejected from the study according to specific transport and

diffusion criteria as described in Results. Retrogradely-

filled cells were identified and manually counted in

successive 1.0 mm fields at 125x. Cells which demonstrated

non-specific fluorescence when exposed to rhodamine (510-560

nm) or fluorescein (450-490 nm) microscope filters were not

counted. Each field was counted 3 times and the median value

was accepted. All cell counts were corrected according to

classical methods (Abercrombie, '46). Total cell number was

obtained by assuming an average cell diameter of 40 Am for

graft cells and 50 Am for host neurons.

The distribution of labeled cells and axons was

determined from drawing tube tracings of darkfield or

brightfield images (HRP and PHA-L) or from photographic

montages of fluorescence micrographs (FG). To determine the

distances of anterogradely labeled axonal projections, a

digitizing tablet and morphometry software (Videoplan;

Kontron, FRG) was calibrated for the appropriate

magnification. Measurements were taken from the drawing tube

illustrations or photomicrographs. Similar methods were used

to document the distances between the injection site and the

outermost zone of tracer diffusion as well as the

relationship of these regions to the host-graft interface.

Immunocytochemical staining and analysis of GFAP

A series of sections (240 gm apart) from recipients with

FG or PHA-L injections was incubated in rabbit polyclonal

antiserum produced against GFAP (gift of Dr. Lawrence F. Eng,

VA Medical Center, Palo Alto, CA). The antiserum was diluted

1:1200, and sections were incubated overnight at 40C.

Detection of the primary antibody was performed according to

the peroxidase anti-peroxidase method (Sternberger, '76) as

described below.

Tracings of the rostral and caudal interface regions for

each section were made using a drawing tube, delineating the

regions containing dense GFAP staining between host and graft

tissue. For specimens with FG injections, the lengths of the

interface and the regions containing glial scar formation

were measured using a digitizing pad and Videoplan software.

The composite Fusion Index (FI) for each interface region was

defined as the average percentage of the interface which was

devoid of dense glial scarring (Houle and Reier, '88).

The density of glial staining was determined for both

graft and surrounding host tissues in 7 recipients. A

program was developed using the Zeiss IBAS image analysis

system (Kontron, FRG) and a high resolution video camera

(DAGE Inc., CCD71). At a viewing magnification of 125x, four

pairs of images from each GFAP stained section (each pair

including a graft region and host gray matter region) were

digitized and converted to binary images. The first field of

each pair was segmented interactively by the user to

distinguish glial processes from background as described by

(Bjorklund,H. et al., '83). Based upon the assumption that

non-specific staining was consistent within each section, the

segmentation settings used for this first image were retained

for the second image of the pair. The percent area occupied

by glial profiles was calculated for each field, and values

were obtained for the average glial density within the graft

and host gray matter regions as well as the ratio of

graft/host glial density for each section. Statistical

comparison of graft and host glial density was performed

using the direct-difference Student's t test for paired

samples (Spence et al., '83).

Immuno-staining For Specific Populations of Host Fibers

Immunohistochemical procedures

Adjacent series of sections from 24 transplant recipients

were stained with polyclonal antisera raised against

neurotransmitters, synthetic enzymes, or peptides found in

specific populations of host fibers (See Table 4-1). Of

this group, 6 recipients were selected from tracer specimens

with unacceptable or failed injections. All of the

recipients were perfused as described above with fixative

containing 4.0% paraformaldehyde and 0.25% glutaraldehyde in

0.1 M Sorenson's phosphate buffer (pH 7.4). Sections

containing the graft and surrounding host spinal cord were

cut at 40 jm on a Vibratome and stored in 0.1 M phosphate

buffered saline prior to staining.


Antisera Source Antibody dilution

5-HT Incstar Corp. 1:3000 overnight
OX Incstar Corp. 1:5000 overnight
TH Eugenetech,Inc. 1:750 overnight
CGRP Peninsula Labs 1:12000 36 hours

Immunocytochemical staining was performed on adjacent
series of sections using antisera directed against
the following:
5HT-serotonin; OX-oxytocin; TH-tyrosine hydroxylase.
CGRP-Calcitonin gene-related peptide

Free floating sections were processed using the PAP

immunocytochemical procedure (Sternberger, '76). The primary

antisera used in this study were all raised in rabbit (Table

4-1) and diluted in high salt buffer containing 0.3% Triton

X-100 (THSB). After the sections were removed from the

primary antibody solution, they were washed 3 times in THSB,

incubated in rabbit anti-goat IgG at 1:20 for 45 min to 1 hr,

washed again in THSB, incubated in Rabbit PAP 1:50-1:200 for

30 min, and finally washed in phosphate buffered saline (PBS)

or ammonium phosphate buffer. The peroxidase was visualized

with 0.05% DAB and 0.003% H202. Staining of fibers with

anti-5HT and anti-TH were done in the presence of 0.001%

nickel ammonium phosphate to produce a black reaction

product. Sections were then mounted on gelatin coated

slides. Some slides were counterstained with 1.0% Neutral

Red or 0.1% Cresyl Violet to reveal nearby cytoarchitecture.

The specificity of antibody staining was evaluated by the

morphological distribution of labeled cells and fibers in the

normal spinal cord.

Analysis of immunocytochemical results

Sections were first examined for labeled cells within the

graft. Selected sections were then photographed on an

Axiophot microscope or drawn using a Zeiss microscope with

2.5x or 40x objective and 12.5 x ocular and 1.0 mag drawing

tube. The distances of fiber ingrowth and qualitative

comparisons of the regions of graft area occupied by stained

fibers were determined from the drawings or photos.

General Transplant Characteristics

Viable transplants were present in 92% of the recipient

rats. Nearly all of the grafts filled the lesion cavity and

showed gross apposition with the surrounding host tissues.

The cellular organization and variability within the host-

graft interface was similar to that described in previous

studies from this laboratory (Reier et al., '86a; Houle and

Reier, '88). With regard to the interface, those sections

counterstained with Cresyl Violet or Neutral Red often

contained regions between the host and graft tissue that were

occupied by small, densely packed cell nuclei resembling

glial cells. In contrast, other regions of the interface

were devoid of an obvious cellular boundary between the two

tissues. In these more integrated areas, the only

distinction between host and graft was a transition in the

general cytoarchitectural organization. A similar range of

host-graft fusion was observed in GFAP stained sections.

Neuroanatomical Tracing With HRP and WGA-HRP

Infections into transplants

Solutions of HRP and WGA-HRP were injected or applied to

16 transplants (Table 4-2). The surviving grafts were

classified based upon the histological analysis of the

injection and the extent of tracer diffusion. The injection

sites were examined under darkfield illumination, and the

extent of each was defined as the area containing a purple-

opaque core and the entire surrounding region of orange TMB

reaction product (Fig. 4-1 a,b).

The injection site was restricted solely to the

transplant in six of the recipients (Table 4-2; Groups A, B).

The two specimens classified into Group A contained the

smallest injection sites, which extended less than 0.5 mm in

maximum diameter. While no labeled cells or axons were

observed in the host spinal cord, these small injections

illustrated the presence of intrinsic graft projections (Fig.

4-la). The majority of retrogradely labeled cells in these

specimens were located within 0.5 mm of the center of the

injection site; however, additional retrogradely filled

neurons were found throughout the grafts. In the four other

recipients (Table 4-2; Group B), the injection sites were




Groupb Intrin.

Eff.c Aff.d
proj. proj.

HGP6(6 wk.)
HGP7(5 wk.)
HGP2(6 wk.)
HGP8(6 wk.)
HGP3(6 wk.)
HGP11(6 wk)
HGP5(9 wk.)
HGP10(6 wk.)
HG3 (2 mo.)
HGP9(6 wk.)

Tungsten wire
Tungsten wire
Tungsten wire
Tungsten wire

+ Indicates evidence of projections
present in host or graft tissue.

from labeled


a The tracers were applied using one of five procedures
(see Methods).
b Recipients included in analysis were classified according
to the extent of tracer diffusion as follows: A Injection
site < 0.5 mm in diameter and confined to graft; B -
Injection larger than 0.5 mm and confined to graft; C -
Injection site within graft, diffused into or slightly over
c Efferent projections of graft axons. Anterograde axon
label extended into host spinal cord.
d Afferent projections from host neurons. Retrogradely
labeled cells found in host spinal cord.

Figure 4-1. HRP and WGA-HRP tracing revealed intrinsic
interactions and some projections of graft and host axons.

a) Drawing tube tracings of sequential sagittal sections
through a graft with a small (Group A) injection. The center
of the injection site is shown as solid black and the area
containing dense reaction product with no discernable cells
and axons is represented by the hatched region. The area
outlined by a dotted line represents a high density of
labeled cells and axons, and each individual cell within the
graft is indicated by larger dots.
b) Labeled cells and axons were distributed throughout a
transplant (t; HGP11) to the host-graft interface
(arrowheads) following a larger HRP/WGA-HRP injection into
the dorsal region of a graft.
c)Labeled graft axons coursed parallel to this region of the
host-graft interface, but do not penetrate the host spinal
cord (h).
d-g) Axonal projections formed between host and graft
tissues. Graft efferent projections were identified by
retrograde transport into transplant neurons following
injections into the host spinal cord(d) and anterogradely
labeled axons (white arrows) extending into the host spinal
cord after an injection into the transplant (e). f) Example
of short-distance ingrowth of host axons into a transplant
following an injection made 1.9 mm rostral to a graft. g)
Illustration of the potential for greater axonal interactions
following an injection which diffused across the dorsal
region of the host-graft interface (specimen HH5). Note that
retrogradely filled neurons (*) may represent intrinsic
projections labeled by tracer diffusion. However, labeled
fibers can be seen in this ventral section where the
diffusion does not confuse the host-graft border. Axons
extended across the interface region (i.e. arrowhead) between
host (heavily labeled) and graft (lightly labeled) tissues.

Scale in a = 1.0 mm; b,c = 200 Am; d-g = 100 Am.

- r 'I
I -


larger than 0.5 mm in diameter but were still confined to the

transplants. An extensive network of intrinsic graft

projections was again evident. Both labeled cells and axonal

profiles were observed throughout the transplants and up to

the interface in all directions (Fig. 4-1 b,c,e).

The transplants in Group B also suggested the presence

of axonal projections between host and graft tissues.

Similar to findings from preliminary studies (Reier et al.,

'86a), retrogradely filled neurons were occasionally found

within the host spinal cord. Efferent projections from graft

neurons were also observed, as anterogradely-filled axons

could be followed across the interface into the host in two

recipients (Fig. 4-le). Unfortunately, the possibility of

additional labeled axons oriented perpendicular to the plane

of section could not be assessed. In contrast, there were

some regions of each host-graft interface where the two

tissues appeared to be separated by a glial partition. In

these regions axons coursed parallel to the interface, but

they did not extend into the host spinal cord (Fig. 4-1 c).

The recipients in Group C had injection sites which were

not confined to the graft. In each case, however, the outer

zone of TMB reaction product extended beyond either the

rostral or caudal border of the graft, while the opaque

center was confined to the graft. At those regions where the

injection extended over the interface, labeled axons and host

neurons were found within 1.0 mm of the host-graft interface,

but few labeled cells were observed farther away. This

labeling pattern differed from the pattern observed following

placement of HRP and WGA-HRP into the normal spinal cord (see

Menetrey et al., '85).

Injections into the host spinal cord

In the reciprocal experiment, HRP and WGA-HRP injections

were made both rostral and caudal to the transplants. The

details of the post-graft intervals, injection site location,

and distances between the injections and the host-graft

interface are summarized in Table 4-3. The injections were

completely confined to the host spinal cord in seven animals.

Evidence for axonal interactions between host and graft

tissues was obtained from specimens which contained large (>

1.5 mm radius) injections that extended to within 2 mm of the

interface region. Of the seven specimens, four contained

retrogradely- filled neurons within the transplants (Fig. 4-

1 d). The number of labeled neurons in these grafts ranged

from a single cell to over 30 cells, with most of these

located within 1.0 mm of the host-graft interface.

In addition to efferent projections from transplant

neurons, the larger HRP/WGA-HRP applications also labeled

axons from host cells that had extended processes into the

transplants. Evidence of such axonal ingrowth was observed

in 3 of 7 of the recipients. In each case, the interface

between host and graft tissues was readily apparent as a

distinct border between the dense axonal labeling in the host


Code I.S.a

HH3(14 mo.) rost.

HH4(5.5 mo.)rost.

HH6(2 mo) rost.

HH7(2 mo.) rost.

HH8(2 mo.) rost.

HH9(6.5 mo) rost.

HH11(6 mo.) rost.









caud. 6.2


e.-Int. proj.

2.1 +



1.4 +

0.6 +


0.0 +


+ Indicates retrogradely filled cells or anterogradely filled
axons present within the transplant.

a Injection Site. Injection placed either rostral (rost.)
or caudal (caud.) to host-graft interface.
b Measured distance from center of injection site to the
host-graft interface.
c Measured distance from the edge of HRP reaction product
or diffusion to the host-graft interface.
d Efferent projections of transplant neurons. Retrogradely
labeled cells within the graft.
e Afferent projections into the graft. Anterogradely
labeled axons from the host spinal cord.

Dist. b Dist.C

and very sparse axonal profiles in the transplant (Fig. 4-1

f). These ingrowing host fibers were restricted to a

peripheral border of the graft and most terminated within a

half millimeter of the host-graft interface. In contrast,

labeled axons stopped abruptly at the interface in the

remaining recipients.

The possibility of more extensive axonal integration

within the interface region was illustrated in one case in

which an injection had extended over a dorsal region of the

interface. The labeled axons and cell bodies within this

graft were excluded from consideration of host-graft

projections. However, a region ventral to the area of

diffusion was marked by the presence of labeled axons that

could be followed between host and graft tissues (Fig. 4-1

Fluoro-Gold Injections: Distribution of Cells

Injections into the host spinal cord

Fluoro-Gold injections were made into the host spinal

cord of 19 recipients at distances ranging from 1.8 5.0 mm

from the host-graft interface. Injections in seven of the

rats met three criteria: 1) the injections were confined to

the host spinal cord, 2) they showed no evidence of diffusion

into the cerebral spinal fluid (as seen by a high degree of

non-specific fluorescence throughout the spinal cord) or

central canal region, and 3) they were large enough to label

a high percentage of host neurons adjacent to the graft.

Results from these animals are summarized in Table 4-4.

Code I.S. (host) # labeleda Fusionb Glialc
(Post-Graft)(mm) cells Index Ratio
(Interval) in graft (%sd) G/H

FGC-20(8 mo) 3.6 mm rost. 0 47%(13) 2.3 *

FGC-17(9 mo) 4.3 mm rost. 38 19%(11) 1.2
FGC-21(8 mo) 2.8 mm rost. 28 40%(19) 4.7 *
FGC-23(8 mo) 1.5 mm rost. 8 8%(13) 10.0 *

FGC-13(8 mo) 4.4 mm caud. 697 51%(20) 3.1 *
FGC-14(5.5m) 3.6 mm caud. 212 40%(16) 1.2
FGC-30(6 wk) 1.8 mm caud. 439 27%(21) 1.1

a Total number of retrogradely filled cells within
the transplant. Cell counts corrected according to
the method of Abercrombie ('46).
b Composite Fusion Index: (FI) percentage of the
host-graft interface (closest to injection) which is
devoid of dense glial scar (average of 4 10
c Ratio of the percent area occupied by glial
elements. indicates that the glial density within
the graft was significantly higher than that in the
surrounding host tissue (p 0.01).

The interface between the host gray matter and graft

tissue was characterized by a sharp decrease in the density

of labeled neurons between adjoining host and transplanted

tissue (Fig. 4-2 a). In all seven grafts, there were regions

of fusion between host and graft tissues as identified with

GFAP staining (see below; Fig. 4-2 b). Six of the grafts

contained retrogradely labeled cells. These included three

transplants (FGC-17,21,23), each with fewer than 50

retrogradely filled neurons within the graft and three others

(FGC-13,14,30), each with more than 200 labeled cells. In

this sample, the differences between the two groups did not

correspond with the post-grafting interval or the distance

between the injection site and the host-graft interface.

However, more labeled cells were found in the three

recipients with FG placed caudal to the graft than those with

injections placed at rostral levels.

Transplant neurons which projected into the host spinal

cord were distributed throughout the graft tissue (Fig. 4-2

e). Most of these retrogradely labeled neurons were

multipolar and small, measuring 8 20 gm in diameter (Fig.

4-2 c), although larger cells were observed occasionally

(Fig. 4-2 d). Histograms were made for each of these six

grafts to show the distribution of the labeled cells as a

function of distance from the host-graft interface (Fig. 4-

3). In the recipients with few labeled cells (top of Fig. 4-

3), there were more cells within the first millimeter of the

transplant and a decrease in the density of labeled cells

with distance from the host-graft interface. However,

labeled neurons were more evenly distributed in grafts

containing many fluorescent cells, regardless of the absolute

length of the transplants.

Figure 4-2. Identification of graft neurons projecting into
the host spinal cord by retrograde transport of Fluoro-Gold.

a) The interface between host (h) and transplant (t) is
characterized by a marked decrease in density of labeled
b) Section adjacent to that shown in a), after staining with
anti-GFAP. A glial scar is present along the dorsal region
of this host-graft interface (bottom half of figure), while
the host and graft tissues are well fused in the ventral
region (between arrowheads).
c) Higher magnification of typical retrogradely labeled cells
within this graft.
d) Example of an occasional large graft neuron that projected
into the host spinal cord.
e) Photo-montage of a sagittal section from specimen FGC-30
to illustrate the distribution of labeled neurons throughout
a transplant following a FG injection into the host spinal
cord (h). The host-graft interface is marked by a white
dotted line.

Scale in a,b,e = 200 Am; c,d = 100 um.

a ~$

W0 4+ M 10 $1
-i 4a) 0
P0CGa0 0
0 0) 0 4J V
P-4 4a Ca4i. -4
U o V ~4J C
rH 4 4J0C
(1w Q c 4

0 0 0%

Sr.1 0*
0 r 4 44
*0 0

v.e a
Sr .1 4

S 10 40 0
410 4J Z-4

tO C r- 4
0 0 "44

4j 4. 0 44 Q



). -H -I a) 04 *'
0 C C: 4J
*4(0 04 5 1 0
to j O a

0H & 0 Q4 4 1

r*4 t3o .c -H
> Q a > 0,
.r tl : 1 >

c4 x u nc
to tou &-
kP -.4 r. 5) u
0 434J > r. 4)

ru it .) o 3:-







UOw )
W X8

Cc 5
Z n







I i i *

sip:) p;laqeI jo sjaqwnN


I: 3!! *




I 5 i i

Injections of FG into transplants

In ten of the recipients with FG injections into the

grafts, the extent of tracer diffusion was confined to the

transplant. While seven grafts had some cells labeled in the

host spinal cord, the numbers of labeled cells ranged from

few host neurons to more than 70 host spinal cord cells and

200 dorsal root ganglion (DRG) neurons.

Figure 4-4 summarizes the location and extent of the

graft injections and the distribution of labeled host

neurons. Five of these FG injections were < 0.5 mm in

diameter (FGC- 2a, H6, H7, HS, H10; Fig. 4-5 a). In these

specimens, nearly all of the labeled neurons were found

within the graft itself. The intrinsic neurons were

concentrated in the region nearest the injection site.

However, some retrogradely labeled cells were found in all

regions of these transplants (Fig. 4-5 b). Labeled neurons

were also found in the adjacent host spinal cord in two of

these recipients. In both cases, the labeled cells were

located within 0.5 mm of the interface zone which was

adjacent to the injection site, and distributed within the

medial or lateral intermediate gray. The greatest number of

labeled host neurons were found in the recipient with an

injection located within 0.2 mm of the host-graft interface


In the remaining transplants from this group (FGC-

H9,25,26,27,31) the injection site was larger than 0.5 mm in

S0 41 0 A

> >0 0,0
0 0 0 II

w a 4 6 10

*H o
0 0

0 0

0 p
o os H
4WU 4 05 $4 *

4.) fH 4, ) O

*r (0 0 4
U) a o .

0 .4 4.)
00 4 V 04)4

4( .4 M 4

-t0 0 0 4 o
SW000 r

0 3-'-4 k

WQ44 0V Z V kQ -4


0 -J NM <0 Nf.

C '0

L *

co 0
(8. C
(0 0 i
410o 5 in 5 .- L


0 5 "0 4

4,' 0

4- -0
U. L* 4)>
u \ l I 0 0>
0 11 41 \

'up 0

i3!~ .. -ia ^ K s s S s sI'

0 0
0 Go

CL .,

.- &41

0)r ^ i*O
0 i

75. 0
(A Lr

0 Z0" ; 8

.C e WL
0: 50 4.5 &J S g ^ M g fS R
I- 0 l "

diameter, but was still confined to the graft. All of these

recipients contained some retrogradely labeled host neurons.

Again, the greatest numbers of host neurons were found when

the injection site was within 0.5 mm of either the rostral or

caudal interface. In addition, the majority of labeled host

neurons were found immediately adjacent to the transplants.

In the best case (FGC-25; Figs. 4-5 c-i), the FG injection

extended to 0.2 mm from the caudal interface. Retrogradely

labeled neurons were found throughout host spinal cord caudal

to the graft, while few neurons were found rostral to the

graft. Within the sacral spinal cord (Fig. 4-5 h; i.e. 4 -

6 mm away), labeled cells were distributed throughout the

dorsal and intermediate gray regions (laminae I-VII) as well

as regions of the ventral horn. In addition, more cells were

located ipsilateral than contralateral to the graft. In this

animal, a large number of ipsilateral DRG cells were found as

well (Fig. 4-5 i). However, sections from cervical and

thoracic spinal cord as well as brainstem and brain contained

no labeled cells in any of the recipients. The absence of

retrograde labeling of descending fibers was in contrast with

the pattern observed following a similar injection into a

normal rat.

Comparison of FG and HRP/WGA-HRP injections in normal rats

To compare the patterns of retrograde cell labeling using

the two tracers and the present injection techniques, each of

these tracers was also injected into a normal rat at the T13

Figure 4-5. Retrogradely labeled host neurons following FG
injections into transplants.

a,b) Intrinsic labeled neurons following a small FG injection
(< 0.5 mm diameter) into a transplant (t). a) The majority of
labeled neurons were located in the immediate vicinity of the
injection, which was adjacent to the rostral host-graft
interface (arrowheads). Cells were also found at the far end
of the graft (b) at a distance of 3 mm from the injection.

c-i) Distribution of retrogradely labeled cells following a
larger injection (dotted line in c) near the caudal border
(arrowheads) of specimen FGC-25. d) Most fluorescent labeled
cells (white arrows) were found immediately adjacent to the
host-graft interface. Inset: Verification of the graft
border was obtained in each case by viewing the interface
region with darkfield optics, using the location of blood
vessels (*) as landmark points. e) GFAP staining of one
section from this specimen illustrates a high degree of
fusion between host and graft (between arrowheads). f) A
patch of retrogradely labeled neurons found approximately 3
mm caudal to the transplant. g) Transverse section of the
sacral spinal cord contains four retrogradely labeled
neurons. h) Composite drawing of 40 sections from the host
sacral spinal cord 4 6 mm caudal to the transplant. Left
in the figure is ipsilateral to the graft, right is
contralateral. The photograph in g: was obtained from the
region enclosed in the box. i) Labeled dorsal root ganglion
neurons ipsilateral to the transplant.

Scale in a,c,d,e = 200 Am; b,f,g,i = 100 pm.

~- 4; e~4


vertebral level. Histological analyses revealed neurons

within the cervical and thoracic spinal cord, brainstem

nuclei, DRG, and cortex of both animals. Similar to findings

from grafted animals, many more labeled cells were found in

each of these regions in the rat which received the pressure

injection of FG than the rat with the HRP/WGA-HRP injection.

PHA-L Injections: Patterns of Axonal Projections

Further definition of the axonal trajectories of host and

graft neurons was made with iontophoretic injections of PHA-

L placed either into transplants or the adjacent host spinal

cord (Table 4-5). The injection region was easily defined by

the presence of darkly filled perikarya. In some specimens,

the injection site contained only a small number of labeled

neurons (Fig. 4-6 a,b). Larger injection sites extended up

to 0.5 mm in diameter.

Projections of transplant neurons

Of the 13 recipients with PHA-L injections into the

transplants, five had acceptable axonal labeling (Table 4-5,

top). All of these injections revealed an extensive network

of axonal projections within the graft. Axons near the

injection site exhibited abundant branching and the nearby

neurons were surrounded by terminal enlargements. Survey

electron micrographs of one such graft showed labeled

terminal boutons throughout the graft neuropil (Jakeman and

Reier, unpublished observations). While the greatest density



I.S.a Regionb Axonal Projections

PHAL-8 (2 mo.)Graft Ventral

Efferent axons can be traced
across interface. Injury filled
host axons also evident.

PHAL-10(10 wk)Graft

PHAL-23(6 mo) Graft

PHAL-15(6 mo)


PHAL-11(6 wk)

PHAL-21(6 mo)


PHAL-27 (12mo)

PHAL-16(6 mo)




axons in rostral

dorsal horn and intermediate gray
regions of host.

Efferent axons present in
dorsal horn, dorsal tracts and
intermediate gray, and in rostral
intermediate gray regions.

Graft dorsal- Efferent axons innervate
lateral host lateral motoneurons.

Graft caudal No Efferent axons.

Host 0.5 mm Afferent axons extending
caudal < 0.2 mm into graft.

Host 1.5 mm Afferent axons extending
caudal < 0.3 mm into graft.

Host 1.5 mm All 1 axons
rostral interface.



Host 0.5 mm Many cells and axons
caudal throughout graft*, most within
1.0 mm of interface.

2.0 mm No afferent axons.


a Injection site. Iontophoretic injection into
graft or host spinal cord.
b Region of graft with injection or distance and
direction between injection in host spinal cord and
the host-graft interface.
Evidence of both anterograde and retrograde

of fibers was found within the injection region, labeled

axons were found in all areas of the grafts.

Three general axonal projection patterns were seen within

the transplants. Within graft regions containing densely

packed neuronal cell bodies, the labeled axons branched

extensively (Figs. 4-6 b,c). In contrast, where few

perikarya were found, labeled axons remained mostly

unbranched and followed a relatively straight trajectory

(Fig. 4-6 b). Finally, at the interface regions, individual

axons often coursed parallel to the host-graft border, and

occasionally extended into the host neuropil (Fig. 4-6 d).

Labeled axons could be followed into the host spinal cord

in four of these grafts. The pattern of axonal outgrowth was

slightly different for each specimen. In one case (PHAL-8),

the injection was placed ventromedially within the transplant

and resulted in injury to fibers in the ventral white matter

of the host. The appearance of these injury-filled axons was

distinct. These axons were very heavily labeled, and they

exhibited bulbous terminal swellings. Similar profiles were

not found in any of the other animals in this group.

Following injections into two other grafts (PHAL-10, 23),

a high density of labeled axons were seen within the

transplants, especially near the dorsal region of the host-

graft interface. Labeled fibers extended across the

interface and into the adjacent rostral or caudal dorsal horn

(laminae I-III) of the host (Fig. 4-7a).












.r-4 C

C VO ( *0 P
0-I WHO 0 to

rl-I*r r-1 (0 r->
H -4 H ( L)
V 41 0 0

$4 14
O,>H e; XHO

to 040 4

0 to 0 5 4
~O :~
0 0 r.

0 O
4-) 0 4.-)4-)

4*Q *H -iOC~
4 D s o' I 1 *
O Cr0 Q
E-4 to U 0

X3 4J 0K

0C r go9
g0 .,gV 0o

0 10 *44 -4 $4 4 $
4 0 ("- 04 ,q 41
to r 444i~0

04S-4 U'U 0-i
9J *4 > ( *g+X J r6.

"'i V X -4 kW H.
C > a) 0 41-
SO44J.VI4 0C 0 U
11# O15 O

- X U0 1X V 0 0

'U4.) 0'-A 'V
m )- VP ^-r- rC V.

-^ r-l4 4 J

(0 0 0f P3* 'X4 f
0J -r- p JU g: C4 ((j +4 ) V go Q) 4-) 0^x" ,
to 0 4J ^nco*l wr-
r- 4 C- La V3 4 0U rP-4
4j ) 04-1 W 0 to 03 44

(1)0 vM oQ wa m'uil'^

W~~q~T j~~~**
J.~ ~- '~

t,. -~


lh '~


g^- C ~h

0 Sj &~. 4

. *.